C₂-Symmetric Amino Acid Amide-Derived Organocatalysts ^†

Abstract

N-alkylated C₂-symmetric amino acid amide derivatives were shown to catalyse the Michael addition of 2-hydroxy-1,4-napthoquinone to β-nitrostyrene, achieving a maximum ee of 44%. The corresponding trifluoroacetic acid salts also catalysed the aldol reaction between 4-nitrobenzaldehyde and hydroxyacetone, leading to the formation of predominantly syn-aldol products in up to 55% ee. Aspects of the solvent dependence of the aldol reaction and the H-bonding of the catalyst were investigated.

1. Introduction

Organocatalysis of organic reactions continues to be of significant interest, particularly the design and application of novel catalysts to asymmetric C–C bond formation [1,2,3,4,5,6,7,8,9,10]. We recently reported [11,12] the preparation of a range of amino acid-derived guanidines which were shown to have some potential as asymmetric organocatalysts in the Michael reaction. It was, however, apparent that several problems were associated with this work. Our initial L-proline-derived catalysts (for example 4) gave, at best, a modest 56% ee for the formation of the Michael adduct 3 from the addition of 2-hydroxy-1,4-napthoquinone 1 to β-nitrostyrene 2. These catalysts were found to be difficult to prepare as the intermediates in their synthesis were prone to racemization. We also investigated several N-protected C₂-symmetric amino acid-derived catalysts including 5 (18–22% ee over three solvents) and 6 (15–26% ee over two solvents). These were not as successful as catalyst 4 and gave very slow rates of conversion due to low basicity (Scheme 1).

We were able to deduce from crystallographic studies that these catalysts took part in strong intramolecular H-bonding, which may be preventing the desired intermolecular interaction of catalyst with substrate. Our overall goal in this research was to develop a catalyst that takes part in strong associative interactions with the reactants. The various modifications made to our catalyst structures did not lead to any improvement in ee, and it was apparent from X-ray studies that the H-bonding patterns observed are not predictable. This suggested that the ability of the guanidine to form multiple strong H-bonds is not a favorable one and our original goal [11,12] to employ a simplified range of base catalysts might be more advantageous [13,14,15]. This proposition was supported by an investigation in the solid state demonstrating extensive intra- and intermolecular H-bonding abilities of the proline, guanidine, and/or amide functional groups within these organic moieties. Based on these observations, we have investigated the simpler N,N-dimethylated C₂-symmetric amides of general structure 7 (Figure 1), which should have less capability for strong intramolecular H-bonding.

2. Preparation of Michael Catalysts

The catalysts 7a–e were prepared by reaction of the required diamine 8a–e with a 2-fold excess of N,N-dimethylphenylalanine [16], which had been activated by treatment with CDI in DMF (Method A). This gave the required catalysts but in consistently low yield (21–33% over 5 examples), and unfortunately, the catalysts were consistently contaminated with unidentified by-products and required repetitive chromatography to achieve high purity. An alternative method (Method B) was attempted for compound 7f, which reversed the coupling and methylation steps. Thus, the diamine 8f (n = 5) was coupled with a 2-fold excess of Z-l-Phe-OH activated using CDI, following which the intermediate bis-Cbz-protected amine was simultaneously deprotected and methylated in situ using Pd/C in the presence of methanolic formaldehyde solution. This gave 7f in an excellent 76% yield, and the method was next applied to the most successful (vide infra) catalysts 7a and 7b. Unfortunately, the method was not applicable to these compounds as they were insoluble in most of the solvents we employed. Even on prolonged reaction times (7–10 days), the deprotection was slow at 1 atm of hydrogen, and low yields were obtained. Despite this, after simple chromatography, the compounds were obtained in high purity. In order to improve this process for catalyst 7c, the corresponding bis-Cbz-protected amine was deprotected using HBr/AcOH, and the diamine formed methylated in situ using Pd/C in the presence of methanolic formaldehyde solution (Method C). This gave catalyst 7c in 80% yield (4.0 g scale) after a simple work-up (Entry 2(c),

Table 1. Preparation of catalysts 7a–f and 10.

Entry	Catalyst	n	A/%	B/%	C/%	D/%
1	7a	0	27	30	---	---
2	7b	1	23	13	80	---
3	7c	2	33	---	---	---
4	7d	3	32	---	---	---
5	7e	4	31	---	---	---
6	7f	5	---	76	---	---
7	10	---	---	63	---	73
8	11a	0	---	---	---	65
9	11b	1	---	---	---	59
10	12	---	---	---	---	38

). We also attempted to prepare the dimethyl-substituted base 10 from Cbz-l-Phe-OH by Method B and had some measure of success (Entry 7(b), Table 1), obtaining a 63% yield. However, the reaction was capricious, and repetition of the hydrogenation step led to varied results. We repeated the reaction using hydrogenation in the presence of formaldehyde over Raney Ni and obtained a better result and found purification was easier, leading to 10 in 73% yield from the intermediate diamine (Method D, entry 7(d)). This method was utilized for the proline-derived catalysts 11a, 11b, and 12, giving them in 65%, 59%, and 38% yield from the intermediate diamines (Scheme 2, Table 1).

3. Michael Reaction—Catalytic Studies

We initially investigated our previously studied reaction [11,12] of 2-hydroxy-1,4-napthoquinone 1 to β-nitrostyrene 2 leading to the Michael product 3 (Scheme 3,

Table 2. Catalysed Michael reaction between 1 and 2.

Entry	Cat.	n	Time/d	ee	Yield/%
1	7a	0	4	26	89
2	7b	1	4	44	70
3	7c	2	4	36	77
4	7d	3	4	44	92
5	7e	4	4	27	96
6	7f	5	4	19	92
7	10	-	4	38	85
8	11a	0	4	10	39
9	11a	0	14	15	87
10	11b	1	4	9	98
11	12	-	4	15	69

).

The catalysts 7a–f were all successful in this reaction, with the catalysts 7b and 7d both giving 44% ee (Table 2, entries 1–6); however, there was no clear relationship between the proximity of the amides and any improvement in ee. We next tried the dimethyl-substituted catalyst 10, hoping for an improvement [17]; however, there was no appreciable increase in ee (Table 2, entry 7). The reactions involving the three proline-derived catalysts (Table 2, entries 8–11) gave very poor ees (9–15%) in comparison to the previous catalysts. Interestingly, compound 11a gave poor conversion over the standard reaction time (Table 2, entry 8) and required 14 days to achieve comparable yields to the other catalysts (Table 2, entry 9). Additionally, despite numerous efforts to obtain suitable crystals of all the catalysts employed, catalyst 11a was the only compound that gave crystals suitable for X-ray analysis, which were obtained on standing from a dichloromethane solution and were found to crystallise in the orthorhombic P2₁2₁2 space group (see

Table 3. X-ray crystallographic data obtained from 11a.

	11a
Formula	C14H26N4O2
MW	282.388
Crystal System	Orthorhombic
Space group	P21212
a/Å	13.7087(5)
b/Å	11.6393(4)
c/Å	5.0039(2)
α/o	90
β/o	90
γ/o	90
V/Å3	798.42(5)
Z	2
T/K	100(2)
λ/Å	1.54178
Dc/g cm−3	1.175
μ(Mo-Ka)/mm−1	0.645
Meas./indep.(Rint) refl.	7332/1497 (0.0350)
Restraints, Parameters	0.109
wR2 (all data)	0.1843
R1	0.0867
Goodness of fit on F2	1.042

for crystallographic details). There was disorder of the entire bridging diamide units, which was best modelled over two sites (split in 65:35 occupancy). Intramolecular hydrogen bonding is observed between the proline N atoms (N2 and s.e.) and nearby amide NH functional groups at distances of 2.39(4) Å (N1(H1)⋯N2) and 2.47(7) Å (N1B(H1BA)⋯N2; Figure 1). Intermolecular H-bonding is also found between the carbonyl O atoms (O1 and O1B and s.e.) and neighbouring diamide nitrogen protons (H1′ and H1BA′; where ′ = x, y, 1 + z) at distances of 2.06(4) Å (O1⋯N1′(H1); Figure 1) and 2.06(6) Å (O1B^…N1B′(H1BA′)). These intermolecular interactions allow the superimposable alignment of the individual units of 11a along the c unit cell direction (Figure 2 and Figure 3). The 1D hydrogen-bonded rows pack efficiently in three dimensions using the common brickwork motif.

Two points of note from the X-ray structure of 11a are, firstly, the disorder present in the -CH₂CH₂- linker. It is likely this disorder will be present in the longer chain analogue, and this might explain the lower ees achieved by these catalysts. Secondly, the intramolecular H-bonds between N1(H1)⋯N2 of 2.39(4) Å represent a relatively strong interaction and might be expected to be preserved in solution [18,19]. If reactions involving these catalysts are determined by the breaking of this bond, then this might be a significant factor in determining ee and the relatively slow progress of the reaction.

Following this work, we investigated the reaction of β-nitrostyrene 2 with dimethyl malonate using the catalysts 7b and 7f and found that the catalysts were ineffective in this reaction. Both 7b and 7f were able to catalyse the Michael addition of acetylacetone and 1,3-diphenylpropane-1,3-dione to β-nitrostyrene 2; however, the ees for this process were low (15% and 11%, respectively); full details are given in the supplementary information.

4. Conclusions on Michael Additions

We can conclude from these reactions that the dimethylamine/amide-linked catalysts are catalysts for this process, but the ees obtained are only modest and do not achieve levels of asymmetric induction achieved by other catalysts [1,2]. There is possibly a strong intramolecular H-bond, based on previous studies on guanidine-based catalysts, which might need to be broken to allow the reaction to proceed with efficiency [11,12,20]. The shorter chain length catalysts (7a to 7d) appear to give better results. With the catalyst precursors in hand, we next went on to study an aldol-type reaction.

5. Aldol Reactions

We were interested in other potential applications of these catalysts and were intrigued by the work of Zhao and co-workers [21] and the more recent work of Jimeno [22], who reported that the catalysts 13 and 14 catalysed the biomimetic aldol reaction [23,24] between aldehyde 15 and hydroxyacetone 16, leading to the formation of syn-17 and anti-17. Interestingly, the free amine catalyst 13 leads to predominantly anti-17, whilst the salt 14 leads to the syn-17 product. (Scheme 4)

6. Preparation of Aldol Catalysts

The required diamine trifluoroacetate catalysts 19a–f, 21 and 23a,b were prepared from the diamines 8a–f and 9 via the Boc-protected intermediates 18a–f [25,26,27], 20 and 22a,b, [28,29] by treatment with trifluoroacetic acid in chloroform, followed by drying under vacuum for 24 h (Scheme 5).

7. Aldol Reaction—Catalytic Studies

The initial study was the reaction between 15 and 16 (Scheme 4) and employed catalysts 19a–f and 21 and intended to study the relationship between the de (syn- to anti-ratio), ee, and the chain length within the catalysts. A standard set of conditions was employed in the initial reactions, and the results are shown in

Table 4. Aldol reaction of 15 and 16 catalysed by 19a–f, 21 and 23a,b ⁽ⁱ⁾.

Entry	Cat.	T (d)	n	Conversion (%)	syn:anti	Syn ee	Anti ee
1	19a	2	0	58	62:38	50	72
2	19b	2	1	35	63:37	48	46
3	19c	2	2	32	66:34	34	36
4	19d	2	3	32	62:38	32	42
5	19e	2	4	32	68:32	30	38
6	19f	2	5	32	61:39	32	40
7	21	2	--	46	66:34	32	36
8	19a	4	0	63 (ii)	57:43	36	34
9	19b	4	1	63 (ii)	62:38	20	36
10	21	4	--	62 (ii)	63:37	24	32
11	19a	4	0	80 (ii),(iii)	51:49	52	60
12	19b	4	1	72 (ii),(iii)	57:43	38	60
13	21	4	--	86 (ii),(iii)	62:38	40	24
14	19a	4	0	90 (iii),(iv)	50:50	36	26
15	19b	4	1	88 (iii),(iv)	56:44	14	39
16	21	4	--	83 (iii),(iv)	55:45	16	14
17	23a	4	0	62	52:48	−30 (v)	−28 (v)
18	23b	4	1	48	52:48	−24 (v)	−24 (v)

(i) General conditions catalyst (0.2 equiv.), 15 (1.0 equiv.), 16 (10 equiv. unless stated), DMF (0.33 M relative to 15), rt. (ii) Compound 15 used at 0.66 M; (iii) 20 equiv. of 16; (iv) Reagent 16 used as solvent (0.33 M relative to 15); (v) Opposite enantiomeric selectivity observed.

(entries 1–7). It was apparent from these results that all 7 catalysts studied gave very similar syn- to anti- ratios, with the syn-17 product as the major diastereoisomer (Table 4, entries 1–7). It was found that the shortest chain length catalyst 19a gave the best levels of conversion and the highest ee, with a 50% ee for syn-17 and a 72% ee for the anti-17 product (entry 1). The ee decreases as the chain length increases, becoming fairly constant at a chain length of 4 or more carbons (Table 4, entries 3–6). The most successful catalysts, 19a, 19b, and 21, were investigated further by increasing the concentration of the reaction and reaction time (Table 4, entries 8–10). This led to an increased conversion; however, the syn:anti selectivity was lower in all cases, as were the ees. We also investigated the effect of increasing the number of equivalents of hydroxyacetone 16 (Table 4, entries 11–13) at the higher reaction concentration. This gave increased conversion (72–86% over the three reactions). In the case of 19a and 19b (Table 4, entries 11 and 12), the syn:anti ratio was much poorer than in the previous examples (Table 4, entries 1 and 2); however, for catalyst 21, this effect was less so (Table 4, entry 13 versus entry 10). The problem associated with increasing the relative amount of hydroxyacetone 16 might stem from the reagent acting as a protic solvent, and we next ran a series of experiments with catalysts 19a, 19b and 21 with 16 as the solvent (Table 4, entries 14–16). As expected, this led to a similar loss of syn:anti selectivity in all cases and much lower ees. This suggests that the protic nature of 16 may be an issue in these reactions. Finally, the proline-derived catalysts 23a and 23b were attempted (Table 4, entries 17, 18). These catalysts gave reasonable conversion but no appreciable syn:anti selectivity and mediocre ee, which was the opposite of that seen for the other catalysts. (similar reversals of selectivity have been reported [22,30] for proline-derived catalysts).

We next investigated the effect of varying solvent in the reaction (

Table 5. Aldol reaction of 15 and 16 catalysed by 19a in differing solvents ⁽ⁱ⁾.

Entry	Solvent	Conversion (%)	syn:anti	syn ee	Anti ee
0	DMF	58	62:38	50	72
1	THF	75	47:53	54	82
2	EtOH	73	47:53	54	60
3	iPrOH	52	40:60	50	70
4	MeCN	44	48:52	50	72
5	DMSO	19	63:37	nd	nd
6	MeOH	18	80:20	nd	nd
7	CH2Cl2	12	33:67	nd	nd
8	(CH2OH)2	10	50:50	nd	nd
9	Water	0	--	--	--

(i) General conditions. Catalyst (0.2 equiv.), 15 (1.0 equiv.), 16 (10 equiv.), solvent (0.33 M relative to 15), rt. nd = not determined.

) and found that under the standard conditions, the best conversions were observed in THF (75%, Table 5, entry 1), ethanol (73%, Table 5, entry 2), acetonitrile (44%, Table 5, entry 3), and propan-2-ol (44%, Table 5, entry 4). It was observed, however, that the syn:anti selectivity of these systems was lower and slightly in favour of the anti-17 product, but the ees of both the syn-17 and anti-17 were of the same magnitude as that observed in DMF (Table 5, entry 0; repeated for convenience). Other solvents gave considerably poorer conversion over 48 h particularly the protic solvents methanol, ethylene glycol, and water (Table 5, entries 6, 8, and 9); however, methanol did give good syn-anti selectivity over 18% conversion).

Based on these experiments, we investigated the use of increased catalyst amount (0.2 equiv.) over a longer time period (96 h) in the best predicted solvents (

Table 6. Aldol reaction of 15 and 16 catalysed by 0.2 equivalents of 19a in differing solvents⁽ⁱ⁾.

Entry	Solvent	Conversion (%)	syn:anti	syn ee	Anti ee
1	DMF	93	57:43	52	79
2	THF	99	45:55	55	76
3	EtOH	92	52:48	46	86
4	MeOH	71	52:48	48	75

(i) General conditions Catalyst (0.2 equiv.), 15 (1.0 equiv.), 16 (10 equiv.), solvent (0.33 M relative to 15), rt.

). In all 4 reactions using 0.2 equivalents of catalyst, the conversion was considerably improved, with over 92–99% conversion for DMF, THF, and EtOH (Table 6, entries 1–3). The syn:anti selectivity was best in the case of DMF (Table 6, entry 1) and was poorer in the case of THF, EtOH, and MeOH (Table 6, entries 2–4). The ees for the anti-17 product remained high at 75–86%, with the ees of the syn-17 fairly consistent at 46–55% ee.

The diastereoselectivity of these reactions is low; however, in some cases, the ees are reasonable. In order to progress the work, we looked at the work of Jimeno, who reported [22] the use of the threonine acyl guanidine catalyst 14 (Scheme 4), which was reported to be a superior catalyst in this reaction in comparison to other amino acid acyl catalysts. It was suggested that a stabilizing hydrogen bonding network amongst the acylguanidinium moiety, the enamine nitrogen, and the aldehyde carbonyl oxygen was an important feature of this catalyst. We thus theorized that the incorporation of a threonine into our catalysts might improve selectivity. We attempted to prepare catalyst 25 via our standard coupling method using N-Boc-l-threonine 24, activation using CDI in dichloromethane, and coupling with ethylene diamine 8a (n = 0). This method, however, led to the formation of a complex product, possibly polymeric in nature. We tried the alternate coupling method reported by Jimeno [22] using EDC.HCl and HOBt, which gave the protected intermediate 25 in 69% yield. This was smoothly deprotected to give 26 using TFA in chloroform (Scheme 6). We applied this catalyst to the aldol reaction and found that over the standard conditions (0.2 equivalents, 4 days), there was an increase in syn:anti selectivity (entry 1); however, the conversion was low, as was the ee of both syn-17 and anti-17. We repeated the reaction at a lower temperature for 8 days (entry 2), and this gave a similar conversion and again poor ees. Reverting to room temperature and allowing the reaction to run for 10 days (

Table 7. Aldol reaction of 15 and 16 catalysed by 26 ⁽ⁱ⁾.

Entry	Time	Conversion (%)	syn:anti	syn ee	Anti ee
1	4	36	64:36	32	6
2	8	40	63:37	22	14
3	10	71	61:39	22	14

(i) General conditions Catalyst 26 (0.2 equiv.), 15 (1.0 equiv.), 16 (10 equiv.), DMF (0.33 M relative to 15), rt.

, entry 3) gave a 71% conversion but only 22% ee for the syn-17 product and 14% ee for anti-17.

These results might indicate that any potential H-bonding interactions to the amino acid portion of catalyst 26 appear to be detrimental to the reaction. Indeed, the proximity of the two hydroxyl groups in catalyst 26 to any intermediates in the reaction might have a similar disruptive effect on conversion to that observed in the reactions performed in water, methanol, and ethylene glycol. The reaction, in agreement with the work of Jimeno [22], does show predominantly syn-selectivity (Table 7).

8. Conclusions on Aldol Reactions

From the aldol reactions investigated, we can conclude that catalyst 19a appears to be the best of our catalysts for this process, leading to a slight bias for the syn-17 product in acceptable ee. In most cases using 19a–f, the l-phenylalanine-derived catalysts, the ees observed were always greater for the anti-17 product. The use of l-proline-derived catalysts had a detrimental effect on selectivity and inverted the enantioselectivity, as observed by others [22,30]. The use of an l-threonine-derived catalyst capable of H-bonding interactions had a detrimental effect on enantioselectivity. Jimeno [11] put forward a transition state 27 (from modelling studies) for his catalyst 14, in which the aldehyde undergoes H-bonding interactions with the guanidinium portion of the enamine intermediate. This explains the syn-selectivity observed; however, no role for the hydroxyl group on the threonine residue was put forward. Barbas et al. [31] put forward a similar transition state 28 for the identical aldol reaction catalysed by l-threonine and tBuO-l-threonine. In these cases, the carboxylic acid of the amino acid is involved in H-bonding to the aldehyde, and there appears to be little difference in the syn-anti selectivity between the l-threonine and tBuO-l-threonine catalysts (Figure 4).

Based on these observations and the poor selectivity observed with the l-threonine-derived catalyst 26, we might speculate that a similar transition state is in operation. However, one complicating factor might be that the catalysts themselves might be present as dienamines in which both amine groups have reacted with the excess of hydroxyacetone. We hope to report further studies on similar catalysts in the future.

9. General Procedures

Unless otherwise noted, reactions were stirred and monitored by TLC. TLC plates were visualized using iodine, phosphomolybdic acid, or under UV light. All anhydrous reactions were conducted under a static argon atmosphere using oven-dried glassware that had previously been cooled under a constant stream of nitrogen. Reagents, dry solvents, and starting materials were purchased from commercial suppliers and used without further purification. Flash column chromatography was performed on Davisil^® silica gel (35–70 microns) with the eluent specified in each case; TLC was conducted on precoated E. Merck silica gel 60 F₂₅₄ glass plates. Unless specified, ¹H and ¹³C NMR spectra were recorded on a Bruker Avance 400 spectrometer (Bruker (UK) Ltd. Coventry, UK) with an internal deuterium lock at ambient temperature at 400/100 MHz with internal references of δ_H 7.26 and δ_C 77.16 ppm for CDCl₃, δ_H 3.31 and δ_C 49.00 ppm for CD₃OD, and δ_H 2.50 and δ_C 39.52 for d₆-DMSO. Infrared spectra were recorded on a Bruker Tensor 37 FT-IR. Mass spectra were determined on a Q Exactive Plus (Thermo Scientific) (Blue Scientific Limited, Cambridge, UK) instrument run with positive electrospray ionization (ESI). Melting points were determined on a Stuart SMP10 (Camlab Ltd. Cambridge UK) apparatus and are uncorrected. Optical rotations were measured in a 0.25 dm cell on an ADP440 polarimeter (Bellingham & Stanley Ltd., Weilheim, Germany). Control reactions between 2-hydroxy-1,4-napthoquinone 2 and 3 under standard conditions and between 4-nitrobenzaldehyde 15 and hydroxyacetone 16 under standard conditions, both in the absence of catalyst were determined to show no appreciable levels of conversion over standard reaction timescales.

10. Experimental

10.1. General Methods for the Preparation of Catalysts

Method A: preparation of 7a–e: CDI (2.3 equiv.) was added to a stirred suspension of N,N-dimethyl-l-phenylalanine (2.0 equiv.) in dry DMF (15 mL per mmol of N,N-dimethyl-l-phenylalanine), and the mixture was stirred with gentle warming until dissolved and stirring continued for 1 h. The mixture was then cooled (0 °C), and the diamine (1.0 equiv.) was added. After stirring for 24 h at room temperature, the reaction mixture was rotary evaporated under reduced pressure then co-evaporated with heptane (3 times) to remove DMF. The products 7a–e were purified by repeated column chromatography (0–20% MeOH/CHCl₃).

(2S,2′S)-N,N′-(ethane-1,2-diyl)bis(2-(dimethylamino)-3-phenylpropanamide) 7a.

Method A: Dimethyl-l-phenylalanine (500 mg, 2.60 mmol), 1,2-diaminoethane (78 mg, 1.30 mmol), CDI (484 mg, 2.3 mmol) gave 7a (144 mg, 27%) as a white solid. Rf 0.32 (10% MeOH in CHCl₃); Mp. 117 °C; [α]_D²⁰ 5.2 (c = 1.0, CHCl₃); δ_H 7.09–7.26 (10H, m, 2 × Ph), 6.18 (2H, br s, 2 × NH), 3.12–3.22 (4H, m, 2 × CH₂), 3.03 (2H, dd, J 8.4, 12.9, 2 × CH), 2.85–2.96 (2H, m, 2 × CH), 2.79 (2H, dd, J 4.7, 13 Hz, 2 × CH), 2.25 (12H, s, 4 × Me); δ_C 172.2, 139.6, 129.6, 128.5, 126.3, 71.1, 42.4, 39.0, 33.9; v_max 3303, 2926, 2859, 2828, 2782, 1648, 1536, 1454, 1238. MS (ESI) m/z 206.1 (100%, [M+2H]²⁺), 411.3, (15%, [M+H]⁺); HRMS (ESI) m/z found 411.2752, C₂₄H₃₅N₄O₂⁺ ([M+H]⁺) requires 411.2755.

(2S,2′S)-N,N′-(propane-1,3-diyl)bis(2-(dimethylamino)-3-phenylpropanamide) 7b.

Method A: Dimethyl-l-phenylalanine (750 mg, 3.89 mmol), 1,3-diaminopropane (144 mg, 1.94 mmol), CDI (882 mg, 5.44 mmol) gave 7b (192 mg, 23%) as a white solid. Rf 0.23 (10% MeOH in CHCl₃); Mp. 106 °C; [α]_D²⁰ +22.4 (c = 1.1, CHCl₃); δ_H 7.24–7.25 (8H, m, 8 × CH), 7.13–7.19 (2H, m, 2 × CH), 6.97 (2H, br t, J 6.2 Hz, 2 × NH) 3.22 (2H, dd, J 5.4, 7.6 Hz, 2 × CH), 3.16 (2H, dd, J 13.3, 7.6 Hz, 2 × CH), 2.92–3.10 (4H, m, 2 × CH₂), 2.89 (2H, dd, J 13.3, 5.4 Hz, 2 × CH) 2.32 (12H, s 4 × Me), 1.43 (2H, pentet, J 6.4 Hz, CH₂); δ_C 172.3, 139.7, 129.4, 128.4, 126.2, 71.2, 42.4, 35.8, 33.6, 29.8; v_max 3310, 2973, 2933, 2775, 1639, 1533, 1494, 1266; MS (ESI) m/z 213.1 (100%, [M+2H]²⁺), (425.3, (98%, [M+H]⁺); HRMS (ESI) m/z found 425.2910, C₂₄H₃₇N₄O₂⁺ ([M+H]⁺) requires 425.2911.

(2S,2′S)-N,N′-(butane-1,4-diyl)bis(2-(dimethylamino)-3-phenylpropanamide) 7c.

Method A: Dimethyl-l-phenylalanine (500 mg, 2.59 mmol), 1,4-diaminobutane (114 mg, 1.29 mmol), CDI (503 mg, 3.10 mmol) gave 7c (189 mg, 33%) as a white solid. Rf 0.26 (10% MeOH in CHCl₃); Mp. 123 °C; [α]_D²⁰ +10.0 (c = 1.0, CHCl₃); δ_H 7.18–7.20 (8H, m, 8 × CH), 7.07–7.13 (2H, m, 2 × CH), 6.80 (2H, br t, J 6.0 Hz, 2 × NH), 3.18 (2H, dd, J 5.4, 7.5 Hz, 2 × CH), 3.02–3.13 (6H, m, 2 × CH, 2 × CH₂), 2.81 (2H, dd, J 5.4, 13.5 Hz, 2 × CH), 2.25 (12H, s, 4 × Me), 1.23–1.30 (4H, m, 2 × CH₂); δ_C 172.1, 139.9, 129.4, 128.4, 126.2, 71.0, 42.3, 38.8, 33.0, 27.0; v_max 3310, 2933, 2865, 2827, 2775, 1642, 1535, 1494, 1248; MS (ESI) m/z 220.1 (4%, [M+2H]²⁺), 439.3, (100%, [M+H]⁺); HRMS (ESI) m/z found 439.3065, C₂₄H₃₉N₄O₂⁺ ([M+H]⁺) requires 439.3068.

(2S,2′S)-N,N′-(pentane-1,5-diyl)bis(2-(dimethylamino)-3-phenylpropanamide) 7d.

Method A: Dimethyl-l-phenylalanine (500 mg, 2.58 mmol), 1,5-diaminopentane (132 mg, 1.29 mmol), CDI (502 mg, 3.10 mmol) gave 7d (191 mg, 33%) as a white solid. Rf 0.23 (10% MeOH in CHCl₃); Mp. 76 °C; [α]_D²⁰ +16.0 (c = 0.96, CHCl₃,); δ_H 7.16–7.19 (8H, m, 8 × CH), 7.08–7.13 (2H, m, 2 × CH), 6.75 (2H, br t, J 6.1 Hz, 2 × NH), 3.05–3.19 (8H, m 4 × CH, 2 × CH₂), 2.81 (2H, dd, J 13.4, 5.3 Hz, 2H, 2 × CH); 2.24 (12H, s, 4 × Me), 1.28–1.36 (4H, m, 2 × CH₂), 1.07–1.14 (2H, m, CH₂); δ_C 172.0, 139.9, 129.3, 128.4, 126.2, 71.0, 42.3, 38.9, 33.0, 29.3, 24.2; v_max 3310, 2940, 1640, 1536, 1454, 1254; MS (ESI) m/z 227.1 (4%, [M+2H]²⁺), 445.3, (100%, [M+H]⁺); HRMS (ESI) m/z found 453.3221, C₂₇H₄₁N₄O₂⁺ ([M+H]⁺) requires 453.3224.

(2S,2′S)-N,N′-(hexane-1,6-diyl)bis(2-(dimethylamino)-3-phenylpropanamide) 7e.

Method A: Dimethyl-l-phenylalanine (496 mg, 1.29 mmol), 1,6-diaminopropane (150 mg, 1.94 mmol), CDI (502 mg, 3.10 mmol) gave 7b (193 mg, 31%) as a white solid. Rf 0.25 (10% MeOH in CHCl₃); Mp. 98 °C; [α]_D²⁰ +12 (c = 1.1, CHCl₃); δ_H 7.15–7.23 (8H, m, 8 × CH), 7.06–7.13 (2H, m, 2 × CH), 6.74 (2H, br t, J 5.2 Hz, 2 × NH), 3.02–3.18 (8H, m, 4 × CH₂), 2.81 (2H, dd, J 13.1, 5.0 Hz, 2 × CH), 2.24 (12H, s, 4 × Me), 1.25–1.35 (4H, m, 2 × CH₂), 1.07–1.21 (4H, m. 2 × CH₂); δ_C 172.1, 140.1, 129.4, 128.4, 126.2, 71.1, 42.4, 39.0, 33.0, 29.5, 26.4; v_max 3306, 2932, 2774, 1643, 1536, 1454, 1249. MS (ESI) 467.3, (100%, [M+H]⁺). HRMS (ESI) m/z found 467.3378, C₂₈H₄₂N₄O₂⁺ ([M+H]⁺) requires 467.3381.

Method B, preparation of 7a, 7b, 7f, and 10.

General method: (i) CDI (2.2 equiv.) was added in portions over 5 min to a stirred solution of Z-Phe-OH (2.0 equiv.) dissolved in CH₂Cl₂ (150 mL). After 10 min, the required amine (1.0 equiv.) was added dropwise as a liquid or dissolved in a small volume of CH₂Cl₂. After 24 h, diethyl ether (100 mL) was added to the solidified mass, which was then filtered through a sinter and the solid washed with 1:1 dichloromethane:diethyl ether until the product was free of imidazole. Drying the product under vacuum gave the required intermediates in 94%, 89%, and 72% yields, respectively. In the case of the intermediate for 10, the product did not precipitate, and thus the reaction was washed with citric acid solution (aq. 10%, 4 × 50 mL), sodium bicarbonate (aq. sat., 2 × 50 mL), and brine (3 × 50 mL), dried (MgSO₄) to give the required product in 96% yield. Data for the first two bis-Cbz-protected intermediates was in full agreement with the literature [32]. (ii) The bis-Cbz-protected intermediate (1 equiv.) was suspended in methanol (100 mL per gram), and formaldehyde solution (aq. 37%, 13 equiv.) and Pd/C (10% w/w, 0.50 g per g of starting material) were added. The mixture was vigorously stirred under a hydrogen atmosphere for 5–14 days. The reaction mixture was filtered through celite, which was washed with methanol and evaporated. The mass obtained was co-evaporated with water (3 × 25 mL) to remove excess formaldehyde, then co-evaporated with toluene (2 × 25 mL) to remove water. The residue was dissolved in chloroform, dried (MgSO₄), filtered, and evaporated to give the crude products. The products 7a, 7b, 7f, and 10 were purified by column chromatography (0–20% MeOH/CHCl₃) and were obtained as solids in 30%, 13%, and 76% yield, respectively. Data for 7a and 7b were identical to that reported above.

(2S,2′S)-N,N′-(heptane-1,7-diyl)bis(2-(dimethylamino)-3-phenylpropanamide) 7f.

Method B: (i) Cbz-l-phenylalanine (2.50 g, 8.35 mmol), 1,7-diaminopropane (530 mg, 4.07 mmol), and CDI (1.50 g, 9.25 mmol) gave dibenzyl ((2S,2′S)-(heptane-1,7-diylbis(azanediyl))bis(1-oxo-3-phenylpropane-1,2-diyl))dicarbamate (intermediate) (2.05 g, 2.90 mmol, 72%) as a white solid. Mp. 195–8 °C; [α]_D²⁰ +17.1 (c 4.1, DMSO); δ_H (d₆-DMSO) 7.95 (2H, br t, J 5.8 Hz, 2 × NH), 7.48 (2H, br d, J 8.7 Hz, 2 × NH), 7.05–7.35 (20H, m, 4 × Ph), 4.89–4.97 (4H, m, 2 × CH₂), 4.16–4.22 (2H, m, 2 × CH), 2.97–3.11 (4H, m, 2 × CH₂), 2.93 (2H, dd, J 13.7, 4.9 Hz, 2 × CH), 2.75 (2H, dd, J 13.7, 10.0 Hz, 2 × CH), 1.16–1.50 (10H, m. 5 × CH₂); δ_C (d₆-DMSO) 171.1, 155.8, 138.1, 137.1, 129.2, 128.3, 128.0, 127.7, 127.5, 126.2, 65.2, 56.3, 38.5, 37.8, 29.0, 28.5, 26.3; v_max 3299, 3062, 3031, 2936, 2854, 1692, 1648, 1532, 1286, 1258, 1239; MS (ESI) m/z (693.4 (100%, [M+H]⁺); HRMS (ESI) m/z found 693.3660, C₄₁H₄₉N₄O₆⁺ ([M+H]⁺) requires 693.3647. (ii) The above Cbz-protected intermediate (1.00 g, 1.44 mmol) and formaldehyde (aq. 37%, 1.51 mL, 18.7 mmol) on hydrogenation (Pd/C 10% w/w, 0.50 g) for 5 days gave 7f (0.53 g, 1.10 mmol) in 76% yield as a white solid. Rf 0.25 (10% MeOH in CHCl₃); Mp. 93–5 °C; [α]_D²⁰ +7.2 (c = 1.0, CHCl₃); δ_H 7.17–7.41 (10H, m, 2 × Ph), 6.91 (2H, br s, 2 × NH), 3.35–3.44 (2H, m, 2 × CH), 3.14–3.26 (6H, m, 2 × CH, 2 × CH₂), 2.95 (2H, dd, J 13.7, 4.7 Hz, 2 × CH), 2.40 (12H, s, 4 × Me), 1.37–1.44 (4H, m, 2 × CH₂), 1.18–1.31 (6H, m. 3 × CH₂); δ_C 171.5, 139.6, 129.4, 128.5, 126.3, 70.8, 42.2, 39.2, 33.1, 29.5, 28.8, 26.8; v_max 3268, 3086, 3028, 2928, 2858, 2826, 2772, 1641, 1557, 1250; MS (ESI) 241.2 (100%, [M+2H]²⁺), 481.4, (4%, [M+H]⁺); HRMS (ESI) m/z found 241.1808, C₂₈H₄₆N₄O₂²⁺ ([M+2H]²⁺) requires 241.1805, m/z found 481.3540, C₂₈H₄₅N₄O₂⁺ ([M+H]⁺) requires 481.3537.

(2S,2′S)-N,N′-(2,2-dimethylpropane-1,3-diyl)bis(2-(dimethylamino)-3-phenylpropanamide) 10.

Method B (i) Cbz-l-phenylalanine (3.69 g, 12.33 mmol), 1,3-diamino-2,2-dimethylpropane 9 (0.60 g, 5.87 mmol), CDI (2.19 g, 13.51 mmol) gave dibenzyl ((2S,2′S)-((2,2-dimethylpropane-1,3-diyl)bis(azanediyl))bis(1-oxo-3-phenylpropane-1,2-diyl))dicarbamate (3.76 g, 5.66 mmol) in 96% yield as a white solid. Mp. 91–94 °C; [α]_D²⁰ -7.3 (c = 4.0, CHCl₃); δ_H 7.04–7.26 (20H, m, 4 × Ph), 7.00 (2H, br s, 2 × NH), 5.51 (2H, br d, J 8.0 Hz, 2 × NH), 4.90–5.02 (4H, m, 2 × CH₂), 4.30–4.47 (2H, m, 2 × CH), 2.79–3.13 (4H, m, 2 × CH₂), 2.54–2.93 (2H, m, 2 × CH₂) 0.55 (6H, s, 2 × CH₃); δ_C 172.0, 156.1, 136.5, 136.2, 129.4, 128.8, 128.6, 128.3, 128.1, 127.1, 67.1, 56.5, 46.1, 34.8, 36.3, 23.5; v_max 3308, 3063, 3031, 2958, 1701, 1654, 1526, 1497, 1234, 1027, 738, 695; MS (ESI) m/z (665.3 (100%, [M+H]⁺); HRMS (ESI) m/z found 665.3336, C₃₉H₄₅N₄O₆⁺ ([M+H]⁺) requires 665.3334. (ii) The above Cbz-protected intermediate (1.59 g, 2.39 mmol) and formaldehyde solution (aq. 37%, 2.6 mL, 35.0 mmol) on hydrogenation (Pd/C 10% w/w, 0.50 g) for 5 days gave 10 (0.688 g, 1.48 mmol) in 63% yield as a white solid. Mp. 81–84 °C; [α]_D²⁰ +22.5 (c = 4.2, CHCl₃); δ_H 7.07–7.23 (12H, m, 2 × Ph, 2 × NH), 3.22 (2H, dd, J 8.1, 5.5 Hz 2 × CH), 3.08 (2H, dd, J 13.7, 8.1 Hz 2 × CH), 2.86 (2H, dd, J 13.7, 5.5 Hz 2 × CH), 2.62–2.72 (4H, m, 2 × CH₂), 2.30 (12H, s, 4 × Me), 0.58 (6H, m. 2 × Me); δ_C 172.0, 139.2, 129.3, 128.4, 126.2, 71.0, 45.6, 42.3, 36.0, 34.1, 23.6; v_max 3304, 3063, 3030, 2959, 1700, 1653, 1526, 1235, 739, 695; MS (ESI) 227.2 (100%, [M+2H]²⁺); HRMS (ESI) m/z found 227.1648, C₂₇H₄₂N₄O₂²⁺ ([M+H]²⁺) requires 227.1648.

Method C: preparation of 7b. (i) Dibenzyl ((2S,2′S)-(propane-1,3-diylbis(azanediyl))bis(1-oxo-3-phenylpropane-1,2-diyl))dicarbamate [32] (8.12 g, 12.75 mmol, prepared as in Method B, part (i)) was added in portions to HBr in AcOH (35% 100 mL, excess) and the mixture stirred for 4 h. Diethyl ether (100 mL) was added and the supernatant liquid decanted from the precipitated solid. This solid was dissolved in water (100 mL) which was extracted with diethyl ether (3 × 50 mL) then basified (NaOH to pH 12) and then further extracted with chloroform (3 × 100 mL). The combined chloroform extracts were dried (MgSO₄), filtered and evaporated to give (2S,2′S)-N,N’-(propane-1,3-diyl)bis(2-amino-3-phenylpropanamide) [32] as a white solid (4.14 g, 11.24 mmol). (ii) This product was dissolved in methanol (50 mL), formaldehyde solution (aq. 37%, 12.1 mL, 0.15 mol) and Pd/C (10% w/w, 1.0 g) were added, and the mixture stirred under a hydrogen atmosphere for 5 days. The mixture was filtered through a celite© pad which was washed with MeOH, and the filtrate evaporated under reduced pressure. After co-evaporation with water (3 × 50 mL) and toluene (2 × 50 mL) the solid residue was dissolved in chloroform, dried (MgSO₄), filtered and evaporated. The residue was dissolved in chloroform (100 mL) and extracted with hydrochloric acid (aq. 2N, 40 mL) and the aqueous phase separated and extracted with chloroform (2 × 50 mL) then basified with NaOH (aq. 2M to pH 14) and extracted with chloroform (3 × 50 mL). These extracts were dried and evaporated to give 7b (4.34 g, 10.22 mmol) in 80% yield as a white solid which gave identical data to that reported above.

Method D: Preparation of 10. (i) CDI (2.55 g, 15.8 mmol) was added to a stirred solution of Boc-l-phenylalanine (3.82 g, 14.39 mmol) in CH₂Cl₂ (50 mL). After 10 min, 1,3-diamino-2,2-dimethylpropane 9 (0.70 g, 6.85 mmol) was added and the mixture stirred for 48 h. The reaction was filtered, and the filtrate washed with citric acid solution (aq. 10%, 4 × 50 mL), sodium bicarbonate (aq. sat., 2 × 50 mL) and brine (3 × 50 mL), dried (MgSO₄), filtered and evaporated to give 20 (2.98 g, 4.99 mmol) in 72% yield as a white solid. Mp. 78–82 °C; [α]_D²⁰ -9.6 (c = 4.0, CHCl₃); δ_H 7.09–7.24 (10H, m, 4 × Ph), 6.75–6.93 (2H, m, 2 × NH), 5.08 (2H, br d, J 7.7 Hz, 2 × NH), 4.26–4.31 (2H, m, 2 × CH), 2.93–3.03 (4H, m, 2 × CH₂), 2.60–2.77 (4H, m, 2 × CH₂) 1.34 (18H, s, 6 × CH₃), 0.62 (6H, s, 2 × CH₃); δ_C 172.2, 155.6, 136.8, 129.4, 128.7, 127.0, 80.2, 56.2, 46.0, 38.4, 36.4, 36.4, 28.4, 23.6; v_max 3305, 3062, 2974, 2930, 1654, 1524, 1496, 1247, 1164, 698; MS (ESI) m/z (579.4 (100%, [M+H]⁺); HRMS (ESI) m/z found 597.3648, C₃₃H₄₉N₄O₆⁺ ([M+H]⁺) requires 597.3647. (ii) Compound 20 (1.00 g, 1.67 mmol) was dissolved in dichloromethane (10 mL), cooled (0 °C) and trifluoroacetic acid (5 mL) was added. After stirring overnight, the mixture was evaporated and dissolved in water (15 mL) following which excess NaOH (aq. 2M) was added, and the mixture extracted with chloroform (3 × 50 mL). The combined organic extract were dried (MgSO₄), filtered and evaporated to give the diamine which was used in the next step without further purification. (iii) The diamine was dissolved in methanol (5 mL per gram) and formaldehyde solution (aq. 37%, 2.50 mL, 33.6 mmol, 13.0 equiv.) and RaneyNi (10% w/w, 0.25 g) were added. The mixture was vigorously stirred under a hydrogen atmosphere for 5 days and the reaction mixture was filtered through celite under a blanket of nitrogen (CAUTION RaneyNi is prone to ignition in oxygen) which was washed with further methanol. The filtrate was evaporated, and the residue obtained was co-evaporated with water (3 × 25 mL) to remove excess formaldehyde then co-evaporated with toluene (2 × 25 mL) to remove water. The residue was redissolved in chloroform, dried (MgSO₄), filtered and evaporated to give 10 (0.55 g, 1.22 mmol) in 73% yield as a gum. Data for 10 was identical to that reported above.

(2S,2′S)-N,N′-(ethane-1,2-diyl)bis(1-methylpyrrolidine-2-carboxamide) 11a.

(2S,2′S)-N,N′-(Ethane-1,2-diyl)bis(pyrrolidine-2-carboxamide) [29] (0.87 g, 3.42 mmol (Prepared by coupling Cbz-l-Pro-OH and ethylene diamine (Method C i), 88%) then HBr/AcOH deprotection (Method D (i), 65%)), formaldehyde solution (aq. 37%, 3.28 mL, 44.0 mmol, 13.0 equiv.) and RaneyNi (10% w/w, 0.25 g) in methanol (5 mL) using Method D (iii) over 7 days (as reported above for 10) gave 11a (0.63 g, 2.23 mmol) in 65% yield as a white solid. Mp. 154–155 °C; Rf 0.23 (10% MeOH in CHCl₃); [α]_D²⁰ −120.4 (c = 4, CHCl₃); δ_H 7.60 (2H, br s, 2 × NH), 3.29–3.46 (4H, m, 2 × CH₂), 3.04–3.12 (2H, m, 2 × CH), 2.88 (2H, dd, J 5.2, 10.2 Hz, 2 × CH), 2.33 (6H, s, 2 × Me), 2.30–2.37 (2H, m, 2 × CH), 2.12–2.24 (2H, m, 2 × CH), 1.63–1.84 (6H, m, 2 × CH, 2 × CH₂); δ_C 175.2, 68.9, 56.7, 41.8, 38.8, 31.1, 24.3; v_max 3275, 2964, 2938, 2872, 3840, 2782, 2763, 1658, 1510, 1457, 1427, 1226, 1048, 745; MS (ESI) 283.2, (10%, [M+H]⁺); HRMS (ESI) m/z found 283.2126, C₁₄H₂₇N₄O₂⁺ ([M+H]⁺) requires 283.2129.

(2S,2′S)-N,N′-(propane-1,2-diyl)bis(1-methylpyrrolidine-2-carboxamide) 11b.

(2S,2′S)-N,N′-(propane-1,3-diyl)bis(pyrrolidine-2-carboxamide) [28] (Prepared by coupling Cbz-Pro-OH and propane-1,3-diamine (Method C (i), 77%) then HBr/AcOH deprotection (Method D (i), 83%). data for dibenzyl 2,2′-((propane-1,3-diylbis- (azanediyl))bis(carbonyl))(2S,2′S)-bis(pyrrolidine-1-carboxylate; gum, [α]_D²⁰ −24 (c = 4.0 CHCl₃); δ_H (d₆-DMSO) 7.83–8.02 (2H, m, 2 × NH), 7.19–7.42 (10H, m, 2 × Ph), 4.94–5.13 (4H, m, 2 × CH₂), 4.06–4.21 (2H, m, 2 × CH), 3.28–3.55 (4H, m, 2 × CH₂), 2.88–3.12 (4H, m, 2 × CH₂), 1.91–1.99 (2H, m, 2 × CH), 1.68–1.91 (6H, m, 2 × CH, 2 × CH₂), 1.37–1.56 (2H, m. CH₂); δ_C (d₆-DMSO) 172.1/172.0/171.8/171.8, 145.1/153.9, 137.0, 128.4/128.3/128.2, 127.8/127.6/127.6/127.5, 127.1/127.0, 65.9/65.8/65.2, 60.3/59.7, 47.1/46.5, 36.1/36.0/36.0/35.8, 31.3/30.2, 29.2, 23.9/23.1; v_max 3292, 3065, 2952, 2879, 1692, 1655, 1532, 1411, 1354, 1239, 1209, 1175, 1118, 1090, 917, 728, 696; MS (ESI) m/z 537.3 (100%, [M+H]⁺); HRMS (ESI) m/z found 537.2708, C₂₉H₃₇N₄O₆⁺ ([M+H]⁺) requires 537.2708.), 0.63 g, 2.35 mmol), formaldehyde solution (aq. 37%, 2.45 mL, 32.6 mmol, 13.9 equiv.) and RaneyNi (10% w/w, 0.25 g) in methanol (4 mL) using Method D (iii) gave 11b (0.41 g, 1.38 mmol) in 59% yield as a white solid. Mp. 99–102 °C; Rf 0.29 (10% MeOH in CHCl₃); [α]_D²⁰ −132.0 (c = 4, CHCl₃); δ_H 7.57 (2H, br s, 2 × NH), 3.17–3.31 (4H, m, 2 × CH₂), 3.03–3.15 (2H, m, 2 × CH), 2.88 (2H, dd, J 4.8, 9.9 Hz, 2 × CH), 2.35 (6H, s, 2 × Me), 2.28–2.41 (2H, m, 2 × CH), 2.13–2.25 (2H, m, 2 × CH), 1.61–1.86 (8H, m, 2 × CH, 3 × CH₂); δ_C 174.8, 69.0, 56.7, 41.8, 35.7, 31.1, 30.1, 24.2; v_max 3290, 2956, 2940, 2926, 2878, 2744, 1651, 1523, 1457, 1152, 774: MS (ESI) 227.2, (10%, [M+H]⁺); HRMS (ESI) m/z found 297.2283, C₁₅H₂₉N₄O₂⁺ ([M+H]⁺) requires 297.2285.

(2S,2′S)-N,N′-(2,2-dimethylpropane-1,3-diyl)bis(1-methylpyrrolidine-2-carboxamide) 12.

(i) Cbz-l-proline (2.28 g, 9.15 mmol, 2.10 equiv.), diamine 9 (0.44 g, 4.31 mmol, 1.0 equiv.) and CDI (1.92 g, 11.84 mmol, 2.6 equiv.) gave dibenzyl 2,2′-(((2,2-dimethylpropane-1,3-diyl)bis(azanediyl)) bis(carbonyl))(2S,2′S)-bis(pyrrolidine-1-carboxylate) (2.44 g, 4.32 mmol) in 95% yield as a gum using Method B, part (i). [α]_D²⁰ -51 (c = 3.0, CHCl₃); δ_H (mixture of rotamers) 8.08/8.13/8.24/8.39/8.52 (2H, br s, NH), 7.04–7.58 (11H, m, 2 × Ph, NH), 4.90–5.19 (4H, m, 2 × CH₂), 4.29–4.38 (2H, m, 2 × CH), 3.39–3.76 (4H, m, 2 × CH₂), 2.61–3.10 (4H, m, 2 × CH₂) 1.86–2.25 (8H, m, 4 × CH₂), 0.61/0.69/0.80/0.88 (6H, br s, 2 × Me); δ_C 172.6/173.0, 154.9/155.6, 149.3, 148.5, 136.3/136.5, 136.2/137.1, 133.7/133.9, 127.7/127.8/128.0/128.4/128.5/128.7/128.8/129.2/130.0/130.5, 119.7, 117.2, 116.0, 69.9, 67.1, 60.8/61.0, 45.2/45.9/46.6/46.9/47.4, 36.6, 31.4, 29.1/29.4, 24.5/24.7, 23.4/23.6. v_max 3292, 3056, 2952, 2880, 1692, 1655, 1411, 1354, 1090, 728, 696; MS (ESI) 565.3, (100%, [M+H]⁺); HRMS (ESI) m/z found 565.3020, C₃₁H₄₁N₄O₆⁺ ([M+H]⁺) requires 565.3021. (ii) The above compound (2.33 g, 4.13 mmol) was deprotected with HBr/AcOH as in method C part (ii) to give (2S,2′S)-N,N′-(2,2-dimethylpropane-1,3-diyl)bis(pyrrolidine-2-carboxamide) (1.00 g, 3.38 mmol) in 82% yield as a gum which was used in the next step without further purification. [α]_D²⁰ -73.8 (c 2.75, CHCl₃); δ_H (d₆-DMSO) 8.13 (2H, t, J 6.8 Hz, 2 × NH), 3.52 (2H, dd, J 5.3, 8.6 2 × CH), 3.33 (2H, br s, 2 × NH obscured), 2.65–2.95 (8H, m, 4 × CH₂), 1.83–2.04 (2H, m, 2 × CH) 1.55–1.74 (6H, m, 2 × CH, 2 × CH₂), 0.74 (6H, s, 2 × Me); δ_C 174.8, 60.3, 46.7, 45.0, 36.5, 30.7, 25.8, 23.2; v_max 3359, 3281, 2959, 2870, 1639, 1523; MS (ESI) 149.1, (100%, [M+H]²⁺); 297.2, (5%, [M+H]⁺); HRMS (ESI) m/z found 297.2282, C₁₅H₂₉N₄O₂⁺ ([M+H]⁺) requires 297.2285. (ii) The above compound (0.88 g, 2.97 mmol), formaldehyde solution (aq. 37%, 4.7 mL, 63.0 mmol, 21.0 equiv.) and RaneyNi (10% w/w, 0.25 g) in methanol (5 mL) using Method D (iii) (as reported above for 10) gave crude material which was purified by column chromatography (0–2% MeOH in chloroform) gave 12 (0.38 g, 1.14 mmol) in 38% yield as a gum. Rf 0.38 (10% MeOH in CHCl₃); [α]_D²⁰ -24 (c = 4, CHCl₃); δ_H 7.78 (2H, t, J 7.1 Hz, 2 × NH), 3.10–3.15 (2H, m, 2 × CH), 3.00 (2H, dd, J 6.9, 13.7 Hz, 2 × CH), 2.96 (2H, dd, J 6.9, 13.7 Hz, 2 × CH), 2.88 (2H, dd, J 5.2, 10.2 Hz, 2 × CH), 2.37 (6H, s, 2 × Me), 2.29–2.35 (2H, m, 2 × CH), 2.13–2.24 (2H, m, 2 × CH), 1.71–1.83 (6H, m, 2 × CH, 2 × CH₂) 0.85 (6H, s, 2 × CH₃); δ_C 175.1, 69.1, 56.7, 45.4, 41.8, 36.8, 31.3, 24.3, 23.6; v_max 3360, 3283, 2959, 2870, 2845, 2785, 1648, 1518, 1451, 1198, 1179, 979; MS (ESI) 163.1 (100%, [M+2H]²⁺); 325.3, (15%, [M+H]⁺); HRMS m/z found 163.1334, C₁₇H₃₂N₄O₂²⁺ ([M+2H]²⁺) requires 163.1335; (ESI) m/z found 325.2596, C₁₇H₃₂N₄O₂⁺ ([M+H]⁺) requires 325.2598.

10.2. Preparation of 19a–f and 17a,b

Compounds 18a–f, 20, 22a,b and 24 were prepared using Method D (i). In the case of 18a–f the compounds were isolated by filtration and trituration with diethyl ether. Compounds 18a–e [27,28,29] and 22a,b [28,29] gave data in accordance with the literature.

Di-tert-butyl ((2S,2′S)-(heptane-1,7-diylbis(azanediyl))bis(1-oxo-3-phenylpropane-1,2-diyl))dicarbamate 18f.

Mp. 132–135 °C; [α]_D²⁰ +15 (c = 4.0, CHCl₃); δ_H (d₆-DMSO, mixture of rotamers) 7.82/7.81–7.90 (2H, br m,/t, J 5.2 Hz, 2 × NH), 7.15–7.27 (10H, m, 2 × Ph) 6.43/6.85 (2H, d/d, J 7.5/8.7 Hz, 2 × NH), 3.95–4.05/4.08–4.13 (2H, m/m 2 × CH), 2.95–3.10 (4H, m, 2 × CH₂), 2.89 (2H, dd, J 4.6, 13.0 Hz, 2 × CH) 2.72/2.61–2.77 (2H, dd/m, J 10.2, 13.0 Hz, 2 × CH), 1.30 (18H, s, 2 × tBu), 1.10–1.42 (10 H, m, 5 × CH₂); δ_C 171.3, 155.2, 138.2, 129.2, 128.0, 126.2, 77.9/78.1, 55.8/57.3, 38.3/38.5, 37.8, 29.0, 28.5, 28.2/27.9, 26.3; v_max 3342, 3319, 2964, 2924, 2852, 1683, 1654, 1252, 1171; MS (ESI) 623.4, (100%, [M+H]⁺); HRMS (ESI) m/z found 625.3960, C₃₅H₅₃N₄O₆⁺ ([M+H]⁺) requires 625.3960.

10.3. Preparation of 19a–f, 21, 23a,b and 25

Compounds 19a–f, 21 and 23a,b were prepared by dissolving the corresponding Boc-protected precursor 18a–f, [25,26,27] 20 or 22a,b [28,29] (0.17 mmol) in chloroform (3 mL) which was cooled (0°C) and trifluoroacetic acid (1–1.5 mL) was added. After stirring at rt for 3 h the reaction was evaporated to dryness and the product dried under vacuum for 24 h. Compound 25 was prepared from diamine 24 (50 mg, 0.17 mmol) by dissolving in chloroform (3 mL) which was cooled (0°C) and trifluoroacetic acid (1 mL) added, then evaporated to dryness under vacuum for 24 h. The catalysts were used directly in the aldol reaction without further purification.

(2S,2′S)-1,1′-(ethane-1,2-diylbis(azanediyl))bis(1-oxo-3-phenylpropan-2-aminium) bistrifluoroacetate salt 19a.

White solid, Mp. 155–159 °C; [α]_D²⁰ +36.1 (c 6.1, MeOH); δ_H (d₆-DMSO) 8.61 (2H, br s 2 × NH), 8.31 (6H, br s, 6 × NH), 7.03–7.51 (10H, m, 2 × Ph), 3.86–4.01 (2H, m, 2 × CH), 2.84–3.11 (8H, m, 4 × CH₂); δ_C 168.1, 158.7 (C, q, ²J_CF 33.6 Hz) 135.1, 129.5, 128.6, 127.2, 116.6 (C, q, J_CF 296.5 Hz), 53.8, 38.0, 37.0; v_max 3321, 3112, 2981, 2867, 2823, 1743, 1671, 1552, 1496, 1058, 655; MS (ESI) 178.1, (100%, [M]²⁺); HRMS (ESI) m/z found 178.1100, C₂₀H₂₈N₄O₂²⁺ ([M]²⁺) requires 178.1101.

(2S,2′S)-1,1′-(propane-1,2-diylbis(azanediyl))bis(1-oxo-3-phenylpropan-2-aminium) bistrifluoroacetate salt 19b.

Gum; [α]_D²⁰ +19.3 (c 10.3, MeOH); δ_H (d₆-DMSO, mixture of rotamers) 8.40 (2H, t, J 5.7 Hz, 2 × NH), 8.03/8.31 (6H, br s, 6 × NH), 7.19–7.32 (10H, m, 2 × Ph), 3.86–4.03 (2H, m, 2 × CH), 2.73–3.07 (8H, m, 4 × CH₂), 1.35/1.86 (2H, 2 × pentet, J 7.6/6.7 Hz CH₂); δ_C 167.8, 158.7 (C, q, ²J_CF 33.6 Hz) 135.1, 129.5, 128.6, 127.2, 116.5 (C, q, J_CF 296.6 Hz), 53.7, 37.2, 36.3/36.4, 25.3/28.3; v_max 3095, 2980, 2873, 1764, 1664, 1499, 1140, 629; MS (ESI) 185.1, (100%, [M]²⁺); HRMS (ESI) m/z found 185.1179, C₂₁H₃₀N₄O₂²⁺ ([M]²⁺) requires 185.1179.

(2S,2′S)-1,1′-(butane-1,2-diylbis(azanediyl))bis(1-oxo-3-phenylpropan-2-aminium) bistrifluoroacetate salt 19c.

Gum; [α]_D²⁰ +46.5 (c 4.0, MeOH); δ_H (d₆-DMSO) 8.40 (2H, t, J 5.7 Hz, 2 × NH), 8.31 (6H, br s, 6 × NH), 7.19–7.35 (10H, m, 2 × Ph), 3.89–4.03 (2H, m, 2 × CH), 2.81–3.14 (8H, m, 4 × CH₂), 1.05–1.26 (4H, m, 2 × CH₂); δ_C 167.8, 158.7 (C, q, ²J_CF 34.7 Hz) 135.2, 129.6, 128.6, 127.2, 116.4 (C, q, J_CF 294.0 Hz), 53.7, 38.3, 37.3, 26.0; v_max 2941, 1776, 1663, 1137, 700; MS (ESI) 192.1, (100%, [M]²⁺); HRMS (ESI) m/z found 192.1256, C₂₁H₃₀N₄O₂²⁺ ([M]²⁺) requires 192.1257.

(2S,2′S)-1,1′-(pentane-1,2-diylbis(azanediyl))bis(1-oxo-3-phenylpropan-2-aminium) bistrifluoroacetate salt 19d.

Gum; [α]_D²⁰ +54.2 (c 4.0, MeOH); δ_H (d₆-DMSO) 8.40 (2H, t, J 5.5 Hz, 2 × NH), 8.27 (6H, br s, 6 × NH), 7.19–7.34 (10H, m, 2 × Ph), 3.86–3.98 (2H, m, 2 × CH), 2.93–3.12 (4H, m, 2 × CH, 2 × CH₂), 2.81–2.92 (2H, m, 2 × CH), 1.12–1.26 (4H, m, 2 × CH₂), 0.90–1.01 (2H, m, CH₂); δ_C 167.5, 158.5 (C, q, ²J_CF 35.1 Hz) 135.0, 129.5, 128.5, 127.2, 116.2 (C, q, J_CF 293.4 Hz), 53.6, 38.5, 37.2, 28.4, 23.5; v_max 2945, 1763, 1664, 1147, 700; MS (ESI) 199.1, (100%, [M]²⁺); HRMS (ESI) m/z found 199.1335, C₂₁H₃₀N₄O₂²⁺ ([M]²⁺) requires 199.1335.

(2S,2′S)-1,1′-(hexane-1,2-diylbis(azanediyl))bis(1-oxo-3-phenylpropan-2-aminium) bistrifluoroacetate salt 19e.

Gum; [α]_D²⁰ +39.0 (c 4.0, MeOH); δ_H (d₆-DMSO) 8.34 (2H, t, J 5.5 Hz, 2 × NH), 8.28 (6H, br s, 6 × NH), 7.20–7.33 (10H, m, 2 × Ph), 3.87–4.00 (2H, m, 2 × CH), 3.04–3.14 (2H, m, 2 × CH), 2.99 (4H, d, J 6.3 Hz, 2 × CH₂), 2.86–2.96 (2H, m, 2 × CH), 1.12–1.32 (4H, m, 2 × CH₂), 0.96–1.12 (4H, m, 2 × CH₂); δ_C 167.9, 158.8 (C, q, ²J_CF 35.0 Hz) 139.8, 129.9, 128.9, 127.6, 117.1 (C, q, J_CF 296.8 Hz), 54.0, 39.1, 37.6, 29.1, 26.4; v_max 2940, 1763, 1663, 1146, 699; MS (ESI) 206.1, (100%, [M]²⁺); HRMS (ESI) m/z found 206.1415, C₂₁H₃₀N₄O₂²⁺ ([M]²⁺) requires 206.1414.

(2S,2′S)-1,1′-(heptane-1,2-diylbis(azanediyl))bis(1-oxo-3-phenylpropan-2-aminium) bistrifluoroacetate salt 19f.

Gum; [α]_D²⁰ +42.9 (c 4.0, MeOH); δ_H (d₆-DMSO) 8.34 (2H, t, J 5.6 Hz, 2 × NH), 8.29 (6H, br s, 6 × NH), 7.20–7.33 (10H, m, 2 × Ph), 3.88–4.00 (2H, m, 2 × CH), 3.06–3.15 (2H, m, 2 × CH), 3.00 (4H, d, J 6.9 Hz, 2 × CH₂), 2.86–2.96 (2H, m, 2 × CH), 1.18–1.35 (4H, m, 2 × CH₂), 1.02–1.17 (6H, m, 3 × CH₂); δ_C 167.6, 158.6 (C, q, ²J_CF 35.8 Hz) 135.1, 129.5, 128.5, 127.2, 116.1 (C, q, J_CF 296.0 Hz), 53.6, 38.7, 37.2, 37.6, 28.7, 28.5, 26.3; v_max 2938, 2863, 1776, 1662, 1141, 700; MS (ESI) 213.2, (100%, [M]²⁺); HRMS (ESI) m/z found 213.1492, C₂₁H₃₀N₄O₂²⁺ ([M]²⁺) requires 213.1492.

(2S,2′S)-1,1′-((2,2-Dimethylpropane-1,3-diyl)bis(azanediyl))bis(1-oxo-3-phenylpropan-2-aminium) bistrifluoroacetate salt 21.

Gum; [α]_D²⁰ +45.8 (c 4.0, MeOH); δ_H (d₆-DMSO) 8.21 (6H, s 6 × NH), 8.22 (2H, t, J 5.4 Hz, 2 × NH), 7.20–7.42 (10H, m, 2 × Ph), 3.99–4.15 (2H, m, 2 × CH), 3.01 (4H, d, J 7.3 Hz, 2 × CH₂), 2.85 (2H, dd, J 6.7, 13.5 Hz, 2 × CH), 2.85 (2H, dd, J 5.7, 13.5 Hz, 2 × CH), 0.56 (6H, s, 2 × Me); δ_C 168.5, 158.8 (C, q, ²J_CF 35.7 Hz), 135.1, 129.6, 128.8, 127.4, 116.1 (C, d, J_CF 269.6 Hz), 53.8, 46.7, 37.4, 36.2, 25.9; v_max 2968, 1764, 1666, 1145, 699; MS (ESI) 199.1, (100%, [M]²⁺); HRMS (ESI) m/z found 199.1334, C₂₃H₃₆N₄O₂²⁺ ([M]²⁺) requires 199.1335.

(2S,2′S)-2,2′-((ethane-1,2-diylbis(azanediyl))bis(carbonyl))bis(pyrrolidin-1-ium) bistrifluoroacetate salt 23a.

Gum; [α]_D²⁰ −22.8 (c 4.0, MeOH); δ_H (d₆-DMSO) 9.81 (2H, br s, 2 × NH), 8.68–8.78 (2H, m, 2 × NH), 8.53 (2H, br s, 2 × NH), 4.09–4.19 (2H, m, 2 × CH), 3.12–3.30 (8H, m, 2 × CH₂), 2.18–2.30 (2H, m, 2 × CH), 1.80–1.92 (6H, m, 2 × CH, 2 × CH₂); δ_C 168.4, 158.8 (C, q, ²J_CF 34.7 Hz), 116.8 (C, d, J_CF 295.1 Hz), 59.1, 45.7, 38.5, 29.5, 23.6; v_max 3290, 3085, 2992, 2958, 1665, 1572, 1169, 1131, 835, 796, 720, 678; MS (ESI) 128.1 (100%, [M]²⁺); HRMS (ESI) m/z found 128.0942, C₁₂H₂₄N₄O₂²⁺ ([M]²⁺) requires 128.0944.

(2S,2′S)-2,2′-((ethane-1,2-diylbis(azanediyl))bis(carbonyl))bis(pyrrolidin-1-ium) bistrifluoroacetate salt 23b.

Gum; [α]_D²⁰ −36.7 (c 4.0, MeOH); δ_H (d₆-DMSO) 9.63–9.96 (2H, br m, 2 × NH), 8.63 (2H, t, J 5.6 Hz, 2 × NH), 8.44–8.59 (2H, br m, 2 × NH), 4.08–4.27 (2H, m, 2 × CH), 3.06–3.33 (8H, m, 4 × CH₂), 2.18–2.39 (2H, m, 2 × CH), 1.75–1.97 (6H, m, 2 × CH, 2 × CH₂), 1.53–1.70 (2H, m, CH₂); δ_C 168.2, 159.0 (C, q, ²J_CF 35.4 Hz), 116.5 (C, q, J_CF 292.8 Hz), 59.2, 45.7, 36.7, 29.8, 28.7, 23.7; v_max 3290, 3092, 2967, 1779, 1665, 1567, 1168, 1132, 835, 797, 721, 704; MS (ESI) 135.1 (100%, [M]²⁺); HRMS (ESI) m/z found 135.1021, C₁₃H₂₆N₄O₂²⁺ ([M]²⁺) requires 135.1022.

Di-tert-butyl ((2S, 2′S, 3R, 3′R)-(ethane-1,2-diylbis(azanediyl))bis(3-hydroxy-1-oxobutane-1,2-diyl))dicarbamate 25.

EDC.HCl (887 mg, 4.63 mmol, 2.30 equiv.), HOBt.H₂O (840 mg, 6.50 mmol, 3.23 equiv.), DIPEA (0.79 mL, 585 mg, 4.52 mmol, 2.25 equiv.) and DMAP (50 mg, 0.41 mmol, 0.20 equiv.) were added sequentially to a cooled (0 °C) solution of l-Boc-Thr 24 (908 mg, 4.14 mmol, 2.06 equiv.) in dry DMF (40 mL) After 1 h, ethylene diamine 8a (n = 0, 121 mg, 0.134 mL, 2.01 mmol, 1.00 equiv.) was added and the resulting mixture was stirred to rt for 48 h. The mixture was diluted with EtOAc (150 mL), washed with NaHCO₃ solution (aq., sat. 2 × 50 mL) and brine (2 × 100 mL), then dried (MgSO₄) and evaporated under reduced pressure. Column chromatography of the crude product (0–2% MeOH in CH₂Cl₂) gave 25 (645 mg, 1.39 mmol) in 69% yield as a white solid. [α]_D²⁰ +14.5 (c 4.0, CHCl₃); Mp. 82–85 °C; δ_H 7.19–7.36 (2H, br m, 2 × NH), 5.72–5.94 (2H, br m, 2 × NH), 4.28–5.39 (2H, br m, 2 × CH), 4.18 (2H, br s, 2 × OH) 4.04 (2H, br d, J 6.0 Hz, 2 × CH), 3.44–3.61 (2H, br m, 2 × CH), 3.19–3.26 (2H, br m, 2 × CH), 1.43 (18H, s, 6 × Me), 1.17 (6H, d, J 6.0 Hz, 2 × Me); δ_C 172.9, 156.5, 80.5, 67.2, 59.8, 39.8, 28.4, 19.3; v_max 3326, 2978, 2935, 1688, 1649, 1497, 1366, 1248, 1161, 910, 729; MS (ESI) 463.3 (100%, [M+H]⁺); HRMS (ESI) m/z found 463.2753, C₂₀H₃₈N₄O₈⁺ ([M+H]⁺) 463.2762.

(2S, 2′S, 3R, 3′R)-1,1′-(ethane-1,2-diylbis(azanediyl))bis(3-hydroxy-1-oxobutan-2-aminium) bistrifluoroacetate 26.

Compound 25 (105 mg, 0.228 mmol) was dissolved in chloroform (4 mL), cooled (0 °C) and trifluoroacetic acid (2 mL) was added. After stirring to rt over 24 h the reaction was evaporated to dryness and the product dried under vacuum for 24 h to give 26 (109 mg) as a gum. This was used directly in the aldol reaction without further purification. [α]_D²⁰ +3.6 (c 2.7, MeOH); δ_H 8.63–8.70 (2H, br m, 2 × NH), 8.01–8.22 (4H, br m, 2 × NH₂), 6.73–8.34 (2H, br s, 2 × OH), 3.85–3.93 (2H, m, 2 × CH), 3.42–3.54 (2H, m, 2 × CH), 3.08–3.31 (4H, m, 2 × CH₂), 1.12 (6H, d, J 6.3 Hz); δ_C 167.2, 158.6 (C, q, ²J_CF 35.8 Hz), 116.3 (C, q, J_CF 296.0 Hz), 65.7, 58.5, 38.2, 20.0; v_max 3254, 3087, 2983, 1662, 1534, 1169, 1181, 1130, 839, 799, 722; MS (ESI) 132.1 (100%, [M]²⁺), 263.2 (25%, [M-H]⁺); HRMS (ESI) m/z found 103.0892, C₂₁H₃₀N₄O₂²⁺ ([M]²⁺) requires 103.0893; m/z found 263.1712, C₂₁H₃₀N₄O₂²⁺ ([M-H]⁺) requires 263.1714.

10.4. General Method for the Reaction of 2-Hydroxy-1,4-napthoquinone 1 with β-Nitrostyrene 2

2-Hydroxy-1,4-napthoquinone 1 (100 mg, 0.574 mmol) and the required catalyst (0.1 equiv.) were dissolved in the requisite solvent (10 mL) and cooled (−20 °C). β-Nitrostyrene 2 (128.5 mg, 0.861 mmol, 1.5 equiv.) was then added and the mixture stirred for the required time and temperature. On completion the solvent was evaporated to give a deep red residue which was purified by column chromatography (2–4% EtOAc in petroleum ether to remove excess 2 then CH₂Cl₂) to give 3 as a yellow solid. Enantiomeric excesses were determined on a CHIRALPAK IA column (250 × 4.6 mm) with 90% hexane with 0.1% TFA, 8% ethanol and 2% dichloromethane as the mobile phase detecting at 254 nm. For a 44% ee sample, S enantiomer 18.1 min, R enantiomer 24.2 min, [α]_D²² −12.1 (acetone, c 1.52; lit. [α]_D¹⁷ −44.8 (acetone, c 1.0) [33], lit. [α]_D²⁵ −34 (acetone, c 1.0) [34].

10.5. General Method for the Aldol Reaction between 4-Nitrobenzaldehyde 15 and Hydroxyacetone 16

The catalyst (0.2 equiv) was dissolved in the required solvent, p-nitrobenzaldehyde 15 (1 equiv.) was added and once dissolved hydroxyacetone 16 (10–20 equiv.). The solution was stirred for the required time, diluted with water (250 mL) and the mixture extracted with ethyl acetate (3 × 50 mL). The combined organic layers were dried with magnesium sulphide, filtered and evaporated. Analysis by ¹H NMR gave the conversion and syn:anti ratio (see SI). Purification by column chromatography (eluting with 10% EtOAc in petroleum ether to remove unreacted 15 was followed by 40–60% EtOAc in petroleum ether) combination of the fractions containing 17 was followed by HPLC analysis to determine ee (see SI). Selected data for the syn-17 diol δ_H = 5.20 (1H, d, J 2.7 Hz) and 5.20 (1H, d, J 2.7 Hz) ppm and the anti-17 diol, δ_H = 5.03 (1H, d, J 4.6 Hz) and 4.41 (1H, d, J 4.6 Hz) ppm. HPLC data:²⁰ Diacel chiralpak AD (250 × 4.6 mm), hexane/i-PrOH = 90:10, flow rate 1.0 mL/min, λ = 254 nm: t_R = 18.4 min (Major anti enantiomer, (3S,4S)), t_R = 21.3 min (Minor anti enantiomer, (3R,4R)), t_R = 26.7 min (Minor syn enantiomer (3S,4R)) and t_R = 37.8 min (Major syn enantiomer, (3R,4S)). Phonemenex Lux^® 3 µm Amylose-1 column (150 × 4.6 mm), hexane/i-PrOH = 90:10, flow rate 1 mL/min, λ = 209 nm: t_R = 14.7 min (Major anti enantiomer, (3S,4S)), t_R = 16.0 min (Minor anti enantiomer, (3R,4R)), t_R = 19.4 min (Minor syn enantiomer (3S,4R)) and t_R = 22.9 min (Major syn enantiomer, (3R,4S)).

11. Crystallography

A single colourless needle-shaped crystal of 11a with dimensions 0.340 × 0.025 × 0.020 mm³ was mounted on a Rigaku 007HF diffractometer with HF Varimax confocal mirrors, a UG2 goniometer and HyPix 6000HE detector. The crystal was kept at a steady T = 100(2) K during data collection. Table 3 contains the basic crystallographic data. CCDC2211459 contains supplementary X-ray crystallographic data for 11a. This data can be obtained free of charge via http://www.ccdc.cam.ac.uk/structures/ (accessed on 15 August 2024), or from the Cambridge Crystallographic Data Centre, Union Road, Cambridge, CB2 1EZ; fax (+44) 1223-336-033 or email: [email protected].