Silver comes into play: Henry reaction and domino cycloisomerisation sequence catalysed by [Ag(i)(Pc-L)] complexes

View Article Online

View Journal

RSC Advances

This article can be cited before page numbers have been issued, to do this please use: G. Tseberlidis, M.

Dell'Acqua, D. Valcarenghi, E. Gallo, E. Rossi, G. Abbiati and A. Caselli, RSC Adv., 2016, DOI:

10.1039/C6RA22231E.

This is an Accepted Manuscript, which has been through the

Royal Society of Chemistry peer review process and has been

accepted for publication.

Accepted Manuscripts are published online shortly after

acceptance, before technical editing, formatting and proof reading.

Using this free service, authors can make their results available

to the community, in citable form, before we publish the edited

article. This Accepted Manuscript will be replaced by the edited,

formatted and paginated article as soon as this is available.

You can find more information about Accepted Manuscripts in the

Information for Authors.

Please note that technical editing may introduce minor changes

to the text and/or graphics, which may alter content. The journal’s

standard Terms & Conditions and the Ethical guidelines still

apply. In no event shall the Royal Society of Chemistry be held

responsible for any errors or omissions in this Accepted Manuscript

or any consequences arising from the use of any information it

contains.

www.rsc.org/advances

Page 1 of 34

RSC Advances

View Article Online

Silver comes into play: Henry reaction and domino cycloisomerisation

DOI: 10.1039/C6RA22231E

sequence catalysed by [Ag(I)(Pc-L)] complexes

Giorgio Tseberlidis,^aMonica Dell’Acqua,^bDaniele Valcarenghi,^aEmma Gallo,^aElisabetta Rossi,^b

Giorgio Abbiati^b* and Alessandro Caselli^a*

^aDipartimento di Chimica, Università degli Studi di Milano, via Golgi 19 – 20133 Milano – Italy

^bDipartimento di Scienze Farmaceutiche, Sezione di Chimica Generale e Organica “A. Marchesini”,

Università degli Studi di Milano, Via Venezian, 21 – 20133 Milano – Italy

Corresponding authors: giorgio.abbiati@unimi.it , alessandro.caselli@unimi.it

Graphical Abstract

Abstract:

We report herein the synthesis of new pyridine-containing macrocyclic ligands (Pc-L) bearing a non-

innocent pendant arm, by exploiting both chiral and functional properties of natural amino acids. The

obtained macrocyclic ligands were employed to synthesize well-defined cationic silver(I) complexes

that were shown to be competent catalyst for the Henry (nitroaldol) reaction. Good to excellent yields

and full selectivity in the -nitroalcohol product were obtained starting from electron-poor aromatic

aldehydes or other activated aldehydes such as furfural under mild reaction conditions. The

straightforward synthesis of the macrocyclic ligands starting from cheap commercially available

starting materials allowed for the introduction of a suitable basic functionality into the ligand pendant

arm, thus providing a bifunctional catalyst. Based on our previous experience in [Ag(I)(Pc-L)]

1

RSC Advances

Page 2 of 34

View Article Online

catalysed domino addition/cycloisomerisation reaction of o-alkynylbenzaldheydes and nucleophiles,

DOI: 10.1039/C6RA22231E

the synthesis of isochromenes coupling the Henry reaction and the cycloisomerisation in a single step

was subsequently explored. Although with low selectivity, [Ag(I)(Pc-L)] cationic complexes were

able to promote such a cascade reaction and a possible mechanism based on experimental evidences

has been proposed.

Introduction

Carbon-carbon bond forming reactions, due to their unique capability of generating molecular

complexity, represent a powerful tool in the reserves of a synthetic chemist.^{1, 2}The last decade has

witnessed considerable advances in the catalytic generation and enantioselective addition of carbon

nucleophiles (i.e., enolates³and nitronates⁴) to different types of electrophiles. In this context,

nitroalkanes are important reagents not only due to their propensity to undergo easy -

dehydrogenation but also for their facile interconversion to other organic functional groups.^{5, 6}Even

weak bases are capable to abstract the proton in -position of a nitro group (pK_a~10) and the

nucleophilic attack of the generated nitronate anion to a carbonyl compound to give a -nitro alcohol

is referred as the Henry (or nitroaldol) reaction.⁷Although more than one century old, this reaction is

still to be considered one of the most important example of an atom-economical transformation.^8-10

No need for a stoichiometric amount of a base, metal-catalysts,^11-15enzymes^{16, 17}or organocatalysts^18-

²⁰can efficiently promote the Henry reaction. Among the metal complexes commonly employed as

catalysts, copper(I)^21-30and copper(II)^{11, 31-44}complexes play a prominent role, whereas, to the best

of our knowledge the activity of silver salts and complexes have rarely been tested. In several cases

reported in the literature, the replacement of Cu with Zn gave comparable results but in some of them

a reversed enantioselectivity was observed.^{27, 32, 33, 42, 45}Cu-catalysed reactions are believed to proceed

by a monometallic form of active species whereas there is strong evidence of the involvement of a

multimetallic species as actual catalyst in Zn(II) promoted Henry reactions.²⁶Surprisingly, such a

correlation in reactivity with silver has never been made and the few examples appeared in the

literature report that silver salts either failed in promoting the reaction,^{40, 46}or gave very poor

yields.^{36, 47}

In the past few years, our attention has turned to the introduction of a pyridine moiety into the skeleton

of tetraaza-macrocycles, with the aim to obtain ligands with increased conformational rigidity and

different basicity.⁴⁸The copper(I) complexes of these 12-membered pyridine-containing ligands (Pc-

L, Fig 1) have been successfully employed as catalysts in the enantioselective cyclopropanation of

alkenes.^49-52The same Cu(I)-catalysts have shown good activities in the Henry reaction.²²Among

2

Page 3 of 34

RSC Advances

View Article Online

coinage metals, reports on the catalytic activity of silver complexes are relatively sparse when

DOI: 10.1039/C6RA22231E

compared to the more extensively studied copper and gold.^53-59We have recently reported the full

characterization of [silver(I)(pyridine-containing ligand)] complexes and their organometallic

reactivity,⁶⁰and we have demonstrated their catalytic activity in the regiospecific domino synthesis

of 1-alkoxy-isochromenes under mild conditions⁶¹and in the microwave enhanced A³-coupling

multicomponent reaction.⁶²Compared with simple silver salts, the great advantages of [Ag(I)(Pc-L)]

complexes are their solubility, their enhanced stability and the easiness of handling. Prompted by the

interesting results above mentioned, we were intrigued to check if our well-defined silver(I)

complexes were suitable catalysts also for the Henry reaction. As a result of this study, we were

pleased to find that [Ag(I)(Pc-L)] complexes can actually activate the aldehyde toward the nitronate

nucleophilic attack, in the first example of a silver mediated Henry reaction. Since the Henry reaction,

especially in the asymmetric version, has provided a good platform for testing the dual activation of

metal/base catalysts,⁶³we have modified our ligands in order to attach in the proper position a suitable

base to facilitate the reaction. Moreover, in connection with our ongoing interest in the study of

domino nucleophilic addition/cyclization reaction involving alkynes characterized by the presence of

a proximate nucleophile,^64-69we envisaged to test the reaction on proper bifunctional substrates.

Based on our experience, we chose the o-alkynylarylaldehydes with the ambition to join the Ag

catalysed Henry reaction (by activation of the aldehyde moiety) to the Ag catalysed

cycloisomerisation (by activation of the triple bond) with the aim to obtain in a cascade fashion new

interesting O-heterocycles such as isochromenes.

Figure 1. Pyridine-containing ligand (Pc-L*) and numbering scheme adopted.

Results and discussion

Silver catalysed Henry reaction. At first, we decided to test the ability of silver to promote the Henry

reaction by using simple silver(I) salts such as Ag(OTf) or Ag(BF₄) and the Pc-L silver(I) complex

1⁶⁰(already used in the A³-coupling study), and to compare their reactivity with those of the

corresponding Pc-L copper(I) complex 2⁵⁰(Figure 2). As model reaction, we chose the condensation

between 4-nitrobenzaldehyde 3a and nitromethane (Table 1). To promote the generation of the

3

RSC Advances

Page 4 of 34

View Article Online

nitronate anion, a catalytic amount of trimethylamine (TEA) - equal to the catalyst load_Din_Og_I:-₁₀w_.10a₃s_9/a_Cd_6Rd_Ae₂d_2231E

to the reaction mixture.

Figure 2. Silver (1) and copper (2) PcL-complexes tested in the model Henry reaction.

Table 1. Preliminary study of the silver(I) catalysed Henry reaction between 4-nitrobenzaldehyde 3a

and nitromethane.^a

Entry

Cat.

Solv.

CH₂Cl₂

CH₃NO₂

Tol

Base

TEA

Cs₂CO₃

TEA

Yield (%)^b

X

1

2^c

3^d

4

Ag(OTf)

Ag(BF₄)

-

21

25

35

84

75

85

84

60

85

OTf

BF₄

OTf

1-(13R)

2-(13R)

5

6^e

7

8

9

CH₂Cl₂

a

Reactions were performed with [Ag(I)] (3.2 x 10^-2mmol) in the solvent (5 mL) at a

cat./base/aldehyde/nitromethane ratio of 1:1:10:50 at rt for 20 h; lower catalyst loadings

b

(1 mol%), resulted in very slow reactions.

Isolated yields based on initial 4-

nitrobenzaldehyde; unreacted aldehyde accounted for the rest of the reaction mass balance.

4

Page 5 of 34

RSC Advances

d

^cT = 30 °C, t = 6 h.

Reaction conditions in the absence of metal catalyst:

View Article Online

DOI: 10.1039/C6RA22231E

TEA/aldehyde/nitromethane ratio of 1:10:500 at rt for 20 h. Unreacted 4-nitrobenzaldehyde

did not account for the rest of the reaction mass balance and some unidentified by-products

derived from competitive side reactions were observed. ^eIn the presence of molecular sieves

(4 Å).

Simple silver(I) salts gave results similar to those observed just in the presence of the base, the only

difference being that in the latter case a number of by-products were observed (Table 1, compare

entries 1, 2 and 3). On the other hand, complex 1, which is fully soluble in chlorinated solvents, gave

results comparable to those obtained with the related copper complex 2 (Table 1, entries 4 and 9).

Under these conditions, the reaction gave identical results either employing the preformed complex

1 or the in situ formed 1:1 ligand/Ag(I) complex. We briefly investigated also the role of the

-

counteranion, by changing from OTf^-to BF₄, obtaining a slightly lower yield with the latter (Table

1, entry 5). However, when the same reaction was repeated in the presence of molecular sieves, a

comparable conversion was again obtained (Table 1, entry 6). The use of a large excess of

nitromethane (Table 1, entry 7) did not affect the reaction yield. We previously observed that the

copper Pc-L complexes failed to catalyse the Henry reaction when it was performed in aromatic

hydrocarbons and that Pc-L ligands alone, in the absence of any metal, are not suitable catalysts.²²

On contrary, the silver Pc-L complex 1 demonstrated to be effective also in toluene, giving the desired

nitroalcohol 4a in 60% yield (Table 1, entry 8). Finally, despite the presence of a defined stereocentre

on the ligand, under all conditions tested we ever obtained the nitroalcohol 4a as a racemic mixture.

Optimization of the silver(I) catalysts for the Henry reaction. Having in hand a quite active

catalytic system, we were intrigued to further optimize the ligand features and then to explore the

scope and limitations of the title reaction. The Pc-L’s are easily constructed in good to excellent yields

taking advantage of a 2+2 Richman-Atkins-type coupling,⁷⁰which has been modified in order to

improve chemical yields, as already reported.⁵⁰In order to extend our method to the synthesis of Pc-

L’s bearing a non-innocent pendant arm on N6 (i.e., a pendant with coordinating, basic or acidic

functionalities), we envisioned that natural amino acids (i.e., -alanine and lysine) could be perfect

nucleophilic reaction partners for the ring opening reaction with 2 moles of N-tosyl aziridine to give

the corresponding 4-substituted 1,4,7-triazaheptanes 5a,b, as depicted in Scheme 1.

5

RSC Advances

Page 6 of 34

View Article Online

DOI: 10.1039/C6RA22231E

Scheme 1. Synthesis of amino acid derived 4 substituted 1,4,7-triazaheptanes 5a and 5b.

In particular, -Alanine methyl ester hydrochloride and (S)-N()-Cbz-Lysine methyl ester

hydrochloride were quantitatively obtained by esterification of the corresponding amino acid with

methanol in the presence of trimethylchlorosilane at room temperature.⁷¹Subsequent reaction with

tosyl protected aziridine in a refluxing mixture of CH₃CN/toluene in the presence of TEA resulted in

the formation of desired bis(sulfonamides) 5a-b in quite fair yields, along with the corresponding

mono adducts 6a-b. However, after chromatographic purification, isolated mono(sulfonamides) 6a-

b were easily converted to 5a-b in very good yields (Scheme 1).

The following Richman-Atkins cyclisation with 2,6-pyridinedimethanol 2,6-dimesylate⁶¹yielded to

the 12-membered macrocycles 7a-b in excellent yields (Scheme 2). Catalytic hydrogenation of 7b

allowed the clean and quantitative removal of the Cbz amino-protecting group to afford 7c, which

was converted to macrocycle 7d, containing a basic tertiary amine pendant arm, by reductive

amination in presence of acetaldehyde and sodium cyanoborohydride (Scheme 2). To check the

stability of the original stereocentre of the starting lysine under the reactions conditions, the optical

rotation of the intermediates 5-7b was measured after each reaction step. The values recorded are

consistent with a complete retention of configuration, as confirmed via ¹⁹F NMR analysis of the

Mosher’s amide formation with ligand 7c (see Supporting Information). To further confirm this data,

ligand 7c was reacted in deuterated chloroform with both enantiomers of the -methoxyphenylacetic

6

Page 7 of 34

RSC Advances

View Article Online

acid in the presence of N,N'-dicyclohexylcarbodiimide (DCC) and 4-dimethylaminopyridine

DOI: 10.1039/C6RA22231E

(DMAP). In both cases, a single diastereoisomer was obtained (see experimental section and

supporting information).

Scheme 2. Synthesis of Pc-L’s 7a-d.

Metal complex formation with ligands 7a-d has been performed with both with Ag(OTf) and

Ag(BF₄), by treating at room temperature a solution of the ligand in dichloroethane (DCE) under

nitrogen atmosphere (Scheme 3). After filtering off eventual uncomplexed metal salt, the colourless

solution was concentrated, and the addition of n-hexane favoured the formation of a white precipitate

of [Ag(Pc-L)] complex in good to excellent yields. All complexes have been isolated and fully

characterized. They showed a remarkable stability both in solid and in solution phase and can be

handled in air without decomposition even for a prolonged period.

7

RSC Advances

Page 8 of 34

View Article Online

DOI: 10.1039/C6RA22231E

Scheme 3. Synthesis of silver(I) complexes 8a-d.

The synthesis of these ligands is relatively simple and high-yielding. Moreover, the major advantage

of this synthetic approach is that it allows an easy modification of a key moiety of the macrocycle,

namely the N6 pendant, by exploiting both chiral and functional properties of the natural amino acids.

The new Ag(I) complexes 8a-d, where tested as catalysts in the model Henry reaction between 4-

nitrobenzaldehyde 3a and nitromethane. According to the best results obtained in the preliminary

screening, the reactions where performed in CH₂Cl₂with a ratio cat/base/aldehyde/CH₃NO₂of

1:1:10:50. As during the optimization studies we noticed that adventitious traces of water could in

some cases affect the reproducibility of the results, all new reactions have been carried out in the

presence of activated molecular sieves. The results are reported in Table 2.

Table 2. The model Henry reaction catalysed by the active-pendant arm Ag(I) complexes 8a-d.^a

Entry

Cat.

Base

Solv.

t (h) T (°C) Yield

X

4a (%)^b

1

2

3

OTf

TEA

CH₂Cl₂

Tol

20

40

20

rt

60

15

53

8a

DiPEA

CH₂Cl₂

8

Page 9 of 34

RSC Advances

View Article Online

4

5

OTf

BF₄

OTf

BF₄

OTf

-

DMAP

Morfoline

K₂CO₃

Cs₂CO₃

TEA

CH₂Cl₂

MeNO₂

MeOH

CH₂Cl₂

20

12

6

r_Dt_{OI: 10.1039/C6RA22231E}

53

8a

8b

-

rt

50

6

45

70

67

65

75

90

95

92

75

45

25

55

-

7

rt

8

rt

9

rt

10

11

12

13

14

15

16

17

18

19^c

20^c

21

22

23

TEA

rt

DiPEA

Cs₂CO₃

-

rt

30

rt

3

20

12

Cs₂CO₃

-

30

rt

BF₄

8c

8d

-

rt

-

d

-

rt

-

Cs₂CO₃

-

30

90

64

^aReactions were performed with [Ag(I)] (3.2 x 10^-2mmol) in the solvent (5 mL) at a cat./base/aldehyde/nitromethane

ratio of 1:1:10:50 in the presence of molecular sieves (4 Å). ^bYields based on initial 4-nitrobenzaldehyde calculated

1

via H NMR using 2,4-dinitrotoluene (DNT) as internal standard; unreacted aldehyde accounted for the rest of the

reaction mass balance. ^cThe metal complex 8c is not soluble in the reaction medium: no reaction after 20 h as judged

d

by TLC analysis. Dimethyl acetal derived from the nucleophilic attack of MeO^-on the 4-nitrobenzaldehyde was

recovered in 25% yield (see SI).

Except for complex 8c, which is the only one not soluble in dichloromethane, all other silver

complexes gave the nitroalcohol 4a from moderate to very good yields in the model reaction. Beside

9

RSC Advances

Page 10 of 34

View Article Online

the desired nitroalcohol 4a the starting aldehyde 3a was the only product recovered at the end of the

DOI: 10.1039/C6RA22231E

reaction, thus yields reported are coincident with conversions and a selectivity > 99% was observed.

Silver complex 8a (Table 2, entries 1-8), characterised by the presence of a slightly coordinating N6

pendant arm, gave on the whole results comparable to those observed by using complex 1. Under the

previously optimized reaction conditions, the reaction yield is slightly lower (cfr. Table 1, entry 4 and

Table 2, entry 1). The use of a less polar solvent resulted in a drop of the reaction yield (Table 2, entry

2). Several bases were tested (Table 2, entries 3-7), and best results were obtained with cesium

carbonate (Table 2, entry 7). The counter anion do not seem to influence the reaction yields (Table 2,

1

entry 8), in agreement with the H NMR spectra of the 8a-BF₄and 8a-OTf that do not show any

apparent interaction between the counterion and the metal complex (see S.I.).

Complex 8b, characterized by the presence of a Cbz-protected amino group on the N6 pendant arm,

displayed a closely related behaviour regarding counter anion (Table 2, entries 9 and 10) and base

(Table 2, entries 10-12) effects, but overall, the yields were higher. With catalyst 8b, the best results

were obtained in the presence of 1 eq. of Cs₂CO₃in dichloromethane (Table 2 entries 12-14): after

20 h at rt, 4a was formed in 90% yield (Table 2, entry 12); a rise in temperature to 30 °C resulted in

an increase of yield to 95% (Table 2, entry 13), whereas a reduction of reaction time gave gradually

worse results (Table 2 entries 14-16). As expected, in the absence of a base, poor results were obtained

(Table 2, entry 17). It should be pointed out that Cs₂CO₃alone is able to promote the Henry reaction,

but under the same reaction conditions only a 55% of nitroalcohol 4a was formed (Table 2, entry 18).

In order to avoid the addition of the base as co-catalyst, we tested complexes 8c and 8d, both

containing a basic amine functionality on the N6 active-pendant arm.

As above pointed out, complex 8c is insoluble in dichloromethane, as well as in nitromethane and

failed to give any reaction in such heterogeneous system (Table 2, entries 19 and 20). When methanol

was used as solvent, we observed the formation of the dimethyl acetal derived from the nucleophilic

attack of MeO^-on the 4-nitrobenzaldehyde in 25% yield (Table 2, entry 21).

Under standard conditions, complex 8d, which is fully soluble in dichloromethane, gave the Henry

product 4a in excellent yields (Table 2, entry 22). Moreover, we were pleased to observe that the

presence of a tertiary amino group on the N6 active-pendant made this complex able to promote the

formation of 4a in 64% yields also without any basic co-catalyst in just 12h at 30 °C (Table 2, entry

23). Unfortunately, again, despite the presence of a stereocentre in S configuration close to the amine,

the enantioinduction in the product is very low (5% ee determined by chiral HPLC).

10

Page 11 of 34

RSC Advances

View Article Online

As already pointed out, the presence of NH₂and/or COOH functional groups in the ligands could

DOI: 10.1039/C6RA22231E

allow for the synthesis of di- or oligo-peptides embedding the metal complex. Hopefully, the chirality

imposed by a more extended peptide structure to the whole catalytic system will provide a more

stereoselective transition state for the two incoming reaction partners. Current efforts in our

laboratory are now devoted in this direction with the aim to improve the enantioselective outcome in

this new silver catalysed Henry reaction.

We next explored scope and limitation of the approach (Table 3), employing complex 8b-BF₄under

the best reaction conditions (Table 2 entries 13 and 14). In particular, our interest was to verify the

ability of our catalytic system to promote the reaction of aldehydes of different nature (aryl, heteroaryl

and cycloalkyl) also in the presence of EW or ED groups on the aromatic ring. We decide to perform

all reactions in CH₂Cl₂at 30 °C and stop them after 12h with the aim to compare the results obtained

by changing the electrophilic properties of the aldehyde at fixed reaction time and temperature. It is

well known that the Henry reaction is an equilibrium reaction and temperature plays an important

role. The 12 h reaction time was decided based on a brief kinetic study performed on the model

reaction (Table 2, entries 13-16), which displayed that a plateau in terms of % of conversion of the

starting aldehyde is reached just after 12 h (Figure 3).

100

90

80

70

60

50

40

30

20

0

5

10

15

20

Hours

Figure 3. Conversion of 3a vs time in the Henry reaction catalysed by 8b-BF₄/Cs₂CO₃in CH₂Cl₂at

30 °C.

Good results in terms of yields were obtained when one or more EWGs were present on the aryl ring

(Table 3, entries 1-6 and 12-16). Although not linearly correlated with _paraHammet constants, yields

between 83% and 93% were obtained with aromatic aldehydes having high positive values of para

11

RSC Advances

Page 12 of 34

View Article Online

substituent constant (Table 3, entries 1-5). Slightly lower yields have been observed for para - and

DOI: 10.1039/C6RA22231E

per-fluorobenzaldehydes (Table 3, entries 6 and 12), and neutral benzaldehyde (Table 3, entry 7). On

the other hand, electron-rich aromatic aldehydes failed to give the Henry products (Table 3, entries

8-11). These latter results can be easily rationalised based on the less pronounced electrophilic

character of the aldehyde. The presence of EW groups is well tolerated also in meta and ortho

positions (Table 2, entries 13-16), suggesting that steric hindrance does not limit the transformation.

As observed for electron-rich benzaldehydes, no reaction was observed with aliphatic aldehydes

(Table 3, entry 19) and with cinnamaldehyde (Table 3, entry 20). Among furan, tiophene and pyrrole

carbaldehydes, a good reactivity was observed only in the case of furfural (Table 3, entry 22), the less

“aromatic” among the five-membered heterocycles, in which, probably, the strong inductive EW

effect of the oxygen atom plays a key role in the activation of the adjacent aldehyde functionality. On

contrary, the reactions of thiophene-2-carbaldehyde and N-methylpirrole-2-carbaldehyde failed

(Table 3, entries 21 and 23). In particular, the very poor yield obtained in the reaction of thiophene-

2-carbaldehyde (Table 3, entry 23) can be also related to a plausible strong coordination of the sulphur

atom to the metal centre that inhibits any catalytic activity, whilst in the case of N-methylpyrrole-2-

carbaldehyde (Table 3, entry 21) only a mixture of unidentified by-products, probably derived from

polymerization, were observed.

Next, we briefly explored the reactivity of some aromatic aldehydes with nitroethane, with a

particular regard to the diastereoselective outcome of the reaction. In all cases, almost identical results

were obtained switching the nitro partner from nitromethane to nitroethane, and all the reactions with

electron poor aromatic aldehydes gave the corresponding Henry product in very good yields, although

with very modest syn/anti ratio (Table 3, entries 24-29). Again, we were pleased to verify that

complex 8d-BF₄was able to catalyse the reaction between 4-nitrobenzaldehyde and nitroethane

without the need of any additional base, yielding 4q in a very satisfying 90% yield (Table 3, entry

25).

Table 3. Scope and limitation of the Henry reaction catalysed by complex 8b-BF₄.^a

Entry

1

R¹

R²

H

Product syn/anti^bYield^c(%)

92 (90)

4-NO₂C₆H₄

-

4a

12

Page 13 of 34

RSC Advances

View Article Online

2

3

4-CNC₆H₄

4-CF₃C₆H₄

4-BrC₆H₄

4-ClC₆H₄

4-FC₆H₄

H

-

_D8_OI3_{: 10}(8_.100₃)_9/C6RA22231E

4b

4c

4d

4e

4f

4g

4h

-

93 (92)

85 (82)

83 (80)

70 (64)

60 (55)

15

4

5

6

7

C₆H₅

8^d

9

4-MeC₆H₄

4-Bu^tC₆H₄

4-MeOC₆H₄

4-Et₂NC₆H₄

C₆F₅

n.d.

10

11

12

13

14

15

16

17

18

19

n.d.

-

n.d.

-

72 (68)

70 (66)

97 (95)

72 (65)

70 (68)

10

4i

4j

4k

4l

4m

4n

-

3,5-(CF₃)₂C₆H₄

2-NO₂C₆H₄

2-BrC₆H₄

2,6-Cl₂C₆H₄

2-MeOC₆H₄

3-MeOC₆H₄

Cy

n.d.

-

20

21

H

-

n.d.

-

22

23

H

-

65 (62)

10

4o

4p

24

4-NO₂C₆H₄

CH₃

60:40

95 (90)

90

4q

25^e

13

RSC Advances

Page 14 of 34

View Article Online

26

27

28

29

2-NO₂C₆H₄

4-CF₃C₆H₄

4-ClC₆H₄

C₆H₅

CH₃

45:55

51:49

61:39

55:45

_D9_OI6_{: 10}(9_.103₃)_9/C6RA22231E

4r

4s

4t

93 (89)

80 (75)

75 (60)

4u

a

Reactions were performed with [Ag(I)] (3.2

x a

10^-2mmol) in CH₂Cl₂(5 mL) at

cat./Cs₂CO₃/aldehyde/nitromethane ratio of 1:1:10:50 in the presence of molecular sieves (4 Å) at 30 °C for

12 h. ^bSyn/anti ratio determined by ¹H NMR. ^cYields based on initial aldehyde calculated via ¹H NMR using

2,4-dinitrotoluene (DNT) as internal standard (isolated yields); unreacted aldehyde accounted for the rest of

the reaction mass balance. ^dReaction performed with complex 8a-BF₄as catalyst. ^eReaction performed with

complex 8d-BF₄as catalyst and in the absence of Cs₂CO₃.

Taking into account that the indole nucleus is present in a large number of compounds of biological

and/or pharmaceutical interest, we tested our approach also on isatine (Scheme 4).

Scheme 4. 8b-BF₄catalysed reaction of nitromethane and nitroethane with isatine.

The reaction with nitromethane gave the desired 3-hydroxy-3-(nitromethyl)-1,3-dihydro-2H-indol-2-

one 9a in 80% isolated yield⁷²whereas the reaction with nitroethane gave a 6:4 diastereoisomeric

mixture of 9b with an overall yield of 97% (yield based on initial isatine calculated via ¹H NMR using

2,4-dinitrotoluene as internal standard, see experimental section).⁷³It should be emphasized that this

transformation on isatine opens up new routes for the synthesis of a plethora of interesting oxindole

alkaloids related molecules.⁷⁴

Silver catalysed domino Henry reaction/cycloisomerisation sequence. As mentioned above, we

have recently reported that [silver(I)(Pc-L)] complexes are competent catalysts for the regioselective

synthesis of 3-substituted-1-alkoxyisochromenes starting from 2-alkynylbenzaldehydes in the

presence of alcohols as nucleophiles.⁶¹Having seen their ability in promoting the Henry reaction, we

were thus intrigued to see if it was possible to combine in a single domino sequence the Henry reaction

of 2-alkynylbenzaldehydes and the cycloisomerisation to yield isochromenes 12. A two-step process

14

Page 15 of 34

RSC Advances

View Article Online

that take advantage of a Cu(II) catalysed Henry reaction followed by an Au(I) mediated

DOI: 10.1039/C6RA22231E

cycloisomerisation⁷⁵has recently been reported by Y. Gong and co-workers (Scheme 5).³⁹

Scheme 5. Alternative domino silver(I) catalysed synthesis of 1-isochromenes.

The reaction between 2-[(4-methoxyphenyl)-ethynyl]benzaldehyde 10a and nitromethane was

selected as model reaction for the optimization of reaction conditions. Firstly, we compared the

activity copper(I) catalyst 2 and silver(I) catalyst 1-OTf (10 mol%) in dichloromethane in the

presence of 5 equiv. of nitromethane and 10 mol% of TEA (Table 4, entries 1 and 2). This preliminary

test interestingly revealed that while the copper complex 2 gave selectively the Henry reaction

product 11a in 56% yield, the silver complex 1-OTf was able to promote the domino sequence,

although with low selectivity, yielding the desired 1-isochromene 12a and the nitroalcohol 11a in

equal amount, beside a 12% of unreacted aldehyde 10a and a complex mixture of unidentified by-

products. An increase of the reaction temperature resulted in lower yields and selectivity, with an

increase of by-products formation (Table 4, entries 3 and 4). The addition of an excess of TEAresulted

in the selective formation of the Henry product in 45% yield, but was detrimental for the formation

of the desired product 12a (Table 4, entry 5). As already stated above, while the Henry product can

be formed also in the absence of the silver catalyst (although in poor yields), the presence of the metal

is essential for the formation of the isochromene 12. In fact, in the presence of 1 equiv. of TEA and

without the metal catalyst, the nitroalcohol 11a was selectively obtained in 77% yield (Table 4, entry

6).

The screening of solvent effect (Table 4, entries 7-9) revealed that polar aprotic solvents favor the

formation of the nitroalcohol 11a. The reduction of the equivalent of nitromethane was detrimental

to the reaction outcome (Table 4, entry 10), whereas, when CH₃NO₂was employed as a

reagent/solvent, the formation of 11a (39%) was accompanied by the formation of a discrete amount

of isochromene 12a (29%, Table 4, entry 11), suggesting that an excess of nitroalkane is able to

15

RSC Advances

Page 16 of 34

View Article Online

promote both isochromene and nitroalcohol formation. Interestingly, when simple AgOTf salt was

DOI: 10.1039/C6RA22231E

used no trace of Henry product 11a was observed, and only a little amount of 12a was obtained (Table

4, entry 12).

Thus, a 50 fold excess of the nitromethane (with respect to aldehyde 10a) was used and this seems to

favor the formation of the isochromene product, especially when freshly distilled nitromethane was

used (Table 4, entries 13 and 14).

-

Among the counter anion tested, BF₄and OTf^-displayed very close results, while NTf₂led to slightly

lower overall yields (Table 4, entries 14-16). It is interesting to note that the presence of 4 Å molecular

sieves as water scavenger seems to speed up the formation of the Henry product to the detriment of

isochromene (Table 4, entry 17), and this selectivity became specificity when the reaction was

performed in acetonitrile (Table 4, entry 18).

We tried to improve the formation of the isochromene 12a by using bases with different pK_b, ranging

from organic to inorganic ones (Table 4 entries 19-24), but the more acceptable results remained those

obtained with TEA or with the more sterically demanding diisopropylethyl amine (DiPEA) (Table 4,

entry 19).

Finally, the use of the catalytic systems previously selected for the Henry approach, confirm their

tendency to promote preferentially the formation of the corresponding nitroalcohol (Table 4, entries

25 and 26), also in the absence of the additional base (Table 4, entry 26).

Table 4. Study on the silver catalysed Henry reaction versus cycloisomerisation reaction.^a

Yield

11a^b

Yield

12a^b

MeNO₂

(equiv.)

T

t

Rec.

Entry

Catalyst Base (equiv.)

Solvent

(h)

10a (%)

(°C)

(%)

(56)

(17)

(%)

-

1

2

TEA (1)

50

CH₂Cl₂20

rt

(35)

(12)

2

(17)

1-OTf

16

Page 17 of 34

RSC Advances

View Article Online

DOI: 10.1039/C6RA22231E

3

4

TEA (1)

TEA (5)

TEA^c

50

CH₂Cl₂30

40

60

rt

12

11

6

7

1-OTf

-

CH₂Cl₂30

CH₂Cl₂22

4

5

50

45 (39)

77

-

11 (10)

6^c

rt

-

10

c

7

TEA (1)

DIPEA (1)

DMAP (1)

DBU (1)

NaHCO₃(1)

Cs₂CO₃(1)

K₂CO₃(1)

TEA (1)

-

50

Tol

DMF

THF

20

24

22

30

rt

30

8

11

1-OTf

AgOTf

1-OTf

1-NTf₂

1-BF₄

1-OTf

1-BF₄

1-OTf

1-NTf₂

8b-BF₄

8d

8

53

-

2

6

9

50

46

21

10

11

CH₂Cl₂22

CH₃NO₂20

CH₂Cl₂22

CH₃CN 22

CH₂Cl₂22

rt

3

-

5

11

-

rt

39

29

-

12

-

rt

-

15

13^d

14^d

15^d

16^d

17^d,e

18^d,e

19^d

20^d

21^d

22^d

23^d

24^d

25^e

500

rt

18

24

-

rt

25

33

-

rt

16

29

5

rt

21 (18)

33 (30)

10 (9)

rt

60 (57) 19 (17)

-

rt

96

36

50

11

5

-

27

-

rt

-

rt

4

rt

4

16

65

2

rt

1

rt

72

78

57

60

-

rt

-

2

rt

15

8

26^e

rt

13

a

b

Reactions were performed with [Ag(I)] (2.5 x 10^-2mmol) in the solvent (1.25 mL) at a catalyst/aldehyde ratio of 1:10.

1

Yields based on initial 10a calculated via H NMR using 2,4-dinitrotoluene (DNT) as internal standard; isolated yields in

brackets. Under these conditions, unreacted starting aldehyde did not always account for the rest of the reaction mass balance.

17

RSC Advances

Page 18 of 34

c

-2

View Article Online

In some cases, unidentified by-products derived from competitive side reactions were detected. TEA (2.5 x 10

DOI: 10.1039/C6RA22231E

d

e

mmol)/aldehyde/nitromethane ratio of 1:1:5. Reaction performed with freshly distilled nitromethane. In the presence of

molecular sieves (4 Å).

With the aim to explain the behaviour observed, we made an additional experiment. Isolated 11a was

reacted in toluene at rt in the presence of 10 mol% of the silver(I) complex 1-OTf. After 24 h no

reaction occurred and the TLC analysis showed the presence of unreacted 11a. Upon addition of 10

1

mol% of TEA the mixture was reacted for additional 24 h at rt, then the crude was analysed by H

NMR which reveals the presence of unreacted 11a (44%) along with traces of 2-[(4-methoxyphenyl)-

ethynyl]benzaldehyde 10a ( 5%), isochromene 11a ( 5%) and some unidentified by-products. This

results suggest that the formation of the nitroalcohol 11a (Path B in scheme 6) and the cascade

synthesis of isochromene 12a (Path A in scheme 6) are probably alternative and competitive

pathways. According to reported metal catalysed domino synthesis of isochromenes in the presence

of nucleophiles,^{61, 76}12a is most likely formed by nucleophilic attack of the nitronate anion on a

preformed isochromenilium ion intermediate I (Path A in Scheme 6), while a subsequent silver

catalysed cycloisomerisation of the nitroalcohol 11a (Path C) seems to be most unlikely.⁷⁷Endorsing

this hypothesis, under our reaction conditions we never isolated or detected in the reactions crude the

alternative isobenzofuran isomers obtained by Gong and co-workers³⁹(see scheme 5).

Scheme 6. Alternative/competitive pathways in the reaction of 2-alkynylarylalehydes and

nitroalkanes.

Up to now, any attempt to obtain in an exclusive fashion the desired isochromene 12a meet with

failure, thus we decided to briefly investigate the substrate effect under the best reaction condition

achieved for the synthesis of isochromenes (Table 4, entry 16). The results are reported in Table 5.

18

Page 19 of 34

RSC Advances

a

View Article Online

Table 5. Substrate effect in the divergent silver catalysed Henry versus cycloisomerisation reaction.

DOI: 10.1039/C6RA22231E

Yield

11^b(%) 12^b(%)

Yield

Rec. SM

(%)

Entry SM

1

R¹

X

R²

H

R³t (h)

H

22

11a 18

12a 30

9

11b +

11b’ 70^c12b’ 7 ^d

12b +

2

p-MeO-Ph CH

Me

21

10a

3

4

Me

H

Me

H

22

-

42^e

5

11c 32

11c 60

12c 31

12c 5

p-Me-Ph

p-CF₃-Ph

Me₃Si

CH

10b

10c

10d

5^f

H

9

12 d

6

H

22

11d 45

14

traces

7

8^f

H

22

1

11e 48

11e 83

11f 15

11g 52

11h 68

11h 78

-

45

15

-

9

n-Pr

CH

N

12f 54

10e

10f

10

11

12^f

-

H

x

22

15

-

p-MeO-Ph

N

10g

a

Reactions were performed with [Ag(I)] (2.5

10^-2mmol) in CH₂Cl₂(1.25 mL) at

a

catalyst/TEA/aldehyde/nitromethane ratio of 1:1:10:500. ^bIsolated yields based on initial alkynylbenzaldehyde

10. Under these conditions, unreacted starting aldehyde did not always account for the rest of the reaction mass

c

balance. In some cases, unidentified by-products derived from competitive side reactions were detected.

d

Mixture of two diastereoisomers in 70:30. ratio. Mixture of two diastereoisomers in 75:25 ratio calculated

on the ¹H NMR. ^eIn this case, the isocumarine 13 (29%) was recovered as major by-product. ^fIn the presence

of molecular sieves (4 Å).

19

RSC Advances

Page 20 of 34

View Article Online

All tested o-alkynylarylaldehydes 10 were readily obtained in moderate to excellent yields by

DOI: 10.1039/C6RA22231E

PdCl₂(PPh₃)₂catalysed Sonogashira coupling reactions⁷⁸starting from commercially available 2-

bromo(hetero)arylaldehydes and terminal acetylenes (see experimental section for details).

Compared to the model reaction (Table 5, entry 1) when nitroethane was used under standard

conditions we observed the formation of only a small amount of distereoisomeric isochromenes 12b

and 12’b (not separated), while the Henry products 11b and 11’b were recovered in 70:30

diastereoisomeric ratio in 70% overall yield (Table 5, entry 2). On the other hand, the reaction with

bulkier 2-nitropropane gave the 3-(4-methoxyphenyl)isocumarine 13 (29%) as main reaction by-

product, beside a huge amount of unreacted starting material (42%) (Table 5, entry 3). The formation

of 13 probably derived from the oxidation of the aldehyde 10a to a carboxylic acid, which was

transformed by the alkaline reaction condition in a carboxylate nucleophile able to make an

intramolecular attack to the silver(I) activated triple bond⁷⁹(Scheme 7).

Scheme 7. Plausible path for the formation of isocumarine 13.

The presence of a neutral aryl substituent on the alkynylbenzaldehyde 10b gave an almost equal

amount of nitroalcohol 11c, and isochromene 12c, (Table 5, entry 4). As already observed (Table 4,

entries 16 and 17), the presence of traces of water seems to be important to promote the formation of

the isochromene products 12; when the reaction of 10b was performed under strictly anhydrous

conditions in the presence of 4 Å molecular sieves, the yield of 12c fall down while the yield of 11c

doubled (Table 5, compare entries 4 and 5). The presence of EWG on the aryl moiety was not suitable

for the formation of the isochromene 12d, and the corresponding nitroalcohol 11d was the main

product obtained (Table 5, entry 6). In addition, trimethylsilyl substituted alkynylbenzaldehyde 10d,

gave exclusively the nitroaldol product 11e (Table 5, entry 7), and when the reaction was conducted

in the presence of a water scavenger, the yield of 11e skyrocket to 83 % (Table 5, entry 8). Alkyl

substitution on the triple bond was the only substitution that lead a good selectivity in favour of the

cyclic product (Table 5, entry 9). More basic and coordinating nicotinaldehyde derivatives 10g,h

yielded exclusively to the Henry products 10h and 10i (Table 5, entries 10 and 11). In the latter cases,

the selectivity in favour of the Henry product could be explained by a speeding up of the nucleophilic

20

Page 21 of 34

RSC Advances

View Article Online

attack of the nitronate anion to the aldehyde, promoted by the presence of an additional basic nitrogen

DOI: 10.1039/C6RA22231E

on the pyridine moiety and by the EW activating effect of the electron-poor pyridine on the proximate

aldehyde group. Also in this case, the presence of molecular sieves resulted in higher yields of

nitroalcohol 11i (Table 5, entry 12).

Conclusions

The [Ag(I)(Pc-L)] complexes employed in the present study are suitable catalysts for the Henry

reaction starting from various electron-poor aromatic aldehydes or other activated aldehydes such as

furfural. Isatine can be converted into the corresponding nitroalcohol in excellent yields. Results

obtained compares well with the state of the art and, at least compared to results obtained by us

employing related copper(I) Pc-L complexes,²²better yields in shorter reaction times could be

obtained. Advantages of the silver catalysed nitro-aldol reaction are: i) mild reaction conditions, ii)

good to excellent reaction yields, iii) cleanness of the reaction and iv) total selectivity in nitroalcohol

formation. All the silver(I) complexes reported are stable and can be handled in open-air atmosphere.

Cs₂CO₃demonstrated to be the base co-catalyst of choice in terms of reaction yields and reduced

reaction times. Moreover, a careful design of the ligand has led to the synthesis of complex 8d, which

contains a suitable basic functionality embedded into the ligand pendant, thus providing a bifunctional

catalyst. A major advantage is represented by the straightforward synthesis of ligand 7d that is

obtained in good overall yield starting from cheap starting materials. Further effort in our research

group will be devoted to the structural modification of the ligand in order to control the

enantioselectivity of this useful transformation.

Based on our experience in [Ag(I)(Pc-L)] catalysed domino addition/cycloisomerisation reaction of

2-alkynylbenzaldehydes and nucleophiles, we tried to find a new entry to isochromenes by coupling

the Henry reaction and the cycloisomerisation in a single domino process. We have demonstrated that

even though with strong limitations, this cascade reaction can occur and that the mechanism involves

a preliminary metal catalysed cycloisomerisation of the 2-alkynylbenzaldehyde to form an

isochromenilium intermediate, followed by the attack from the nucleophilic nitronate anion. A

catalytic amount of base is necessary to generate enough nitronate anion in the reaction medium.

Interestingly, this pathway seems to be alternative and competitive with the formation of the

nitroalcohol via the Ag catalysed Henry reaction. Since in our reaction conditions the Henry reaction

is in general faster, this results in a non-selective process, yielding a mixture of nitroalcohols 11 and

isochromenes 12. While is still possible to direct the reaction towards the nitroalcohol formation, a

full selectivity toward the isochromene product has not yet been obtained.

21

RSC Advances

Page 22 of 34

View Article Online

DOI: 10.1039/C6RA22231E

Experimental procedures

General experimental details. All of the reactions that involved the use of reagents sensitive to

oxygen or hydrolysis were carried out under an inert atmosphere. The glassware was previously dried

in an oven at 110 °C and was set with cycles of vacuum and nitrogen. Also syringes, used to transfer

reagents and solvents, were previously set under a nitrogen atmosphere. The syntheses of the silver

complexes were carried out under a nitrogen atmosphere by employing standard Schlenk techniques.

All chemicals and solvents were commercially available and were used after distillation or treatment

with drying agents. The chromatographic column separations were performed by a flash technique,

using silica gel (pore size 60 Å, particle size 230−400 mesh, Merck grade 9385). For TLC, silica was

used on TLC Alu foils with fluorescent indicator (254 nm) and the detection was performed by

irradiation with UV light (λ = 254 nm or 366 nm). ¹H NMR analyses were performed with 200, 300,

400 or 600 MHz spectrometers at room temperature. The coupling constants (J) are expressed in hertz

(Hz), and the chemical shifts (δ) in ppm. ¹³C NMR analyses were performed with the same

instruments at 75.5, 100 or 150 MHz, and attached proton test (APT) sequence was used to distinguish

the methine and methyl carbon signals from those arising from methylene and quaternary carbon

1

atoms. All ¹³C NMR spectra were recorded with complete proton decoupling. The H NMR signals

of the ligand described in the following have been attributed by correlation spectroscopy (COSY) and

13

nuclear Overhauser effect spectroscopy (NOESY) techniques. Assignments of the resonance in C

NMR were made using the APT pulse sequence and heteronuclear single quantum correlation

(HSQC) and heteronuclear multiple bond correlation (HMBC) techniques. Low resolution MS

spectra were recorded with instruments equipped with electron ionization (EI), ESI/ion trap (using a

syringe pump device to directly inject sample solutions), or fast atom bombardment (FAB) (for Pc-L

and metal complexes) sources. The values are expressed as mass−charge ratio and the relative

intensities of the most significant peaks are shown in brackets. UV−vis spectra of the ligand and its

silver complexes were recorded in CH₂Cl₂. Elemental analyses were recorded in the analytical

laboratories of Università degli Studi di Milano. Optical rotations were measured on a Perkin Elmer

instruments model 343 plus; [α]_Dvalues are given in 10^-1deg cm²g^-1. Silver complex 1⁶⁰, N-tosyl-

aziridine and copper complex 2⁵⁰and 2,6-pyridinedimethanol 2,6-dimesylate⁶¹were synthetized as

previously reported.

Synthesis of -alanine methyl ester hydrochloride. TMS-Cl (8.7 mL, 68.6 mmol) was added

dropwise to -Ala (3.06 g, 34.3 mmol). Then MeOH (27.0 mL) was added slowly and the resulting

solution was stirred for 24h at rt. The solvent was evaporated to dryness leading to (2) (4.8 g, >99

22

Page 23 of 34

RSC Advances

1

View Article Online

%) as a white solid. H NMR (300 MHz, D₂O, δ) 3.81 (s, 3H, OCH₃), 3.36 (t, J = 6.4_DH_OIz_{: 1},₀2_.10H₃₉,_/C_C6H_RA₂₂)₂,_231E

2.89 (t, J = 6.4 Hz, 2H, CH₂). The spectral data are consistent with those previously reported.⁷¹

Synthesis of N(α)-Cbz-lysine methyl ester hydrochloride. TMS-Cl (0.9 mL, 7.0 mmol) was added

dropwise to N(α)-Cbz-lysine (0.98 g, 3.0 mmol). Then MeOH (10.0 mL) was added slowly and the

resulting solution was stirred for 24h at rt. The solvent was evaporated to dryness leading to (3) (1.18

1

g, quantitative yield) as a colorless oil. H NMR (300 MHz, D₂O, δ) 7.62 – 7.21 (m, 7H, H_arand

NH₂), 5.20 (s, 2H, CH₂), 4.28 (m, 1H, CH), 3.80 (s, 3H, OCH₃), 3.01 (m, 2H, CH₂), 2.39 (s, 1H, NH),

2.06 – 1.63 (m, 4H, CH₂), 1.49 (m, 2H, CH₂). The spectral data are consistent with those previously

reported.⁸⁰

Synthesis of 5a and 6a. -Alanine methyl ester hydrochloride (0.377 g, 2.7 mmol) was suspended in

CH₃CN (10.0 mL). TEA (0.45 mL, 3.22 mmol) was added, the mixture was warmed at 60°C and half

of a solution of N-tosyl aziridine (1.33 g, 6.74 mmol in 10.0 mL of toluene) was added dropwise in

10 minutes. After one hour was added dropwise the second half of the solution and the mixture was

left to react for 6h at reflux temperature. The reaction solvent was evaporated to dryness and the crude

was purified by flash chromatography (SiO₂, AcOEt/n-hexane 6:4 gradient to AcOEt/ MeOH 20%)

yielding to 6a (378 mg, yield 47%) and 5a (402 mg, yield 30%).

5a: ¹H NMR (300 MHz, CDCl₃, δ) 7.77 (d, J = 8.2 Hz, 4H, H_ar), 7.30 (d, J = 8.2 Hz, 4H, H_ar), 5.50

(m, 2H, NH), 3.74 (s, 3H, OCH₃), 2.88 (m, 4H, CH₂), 2.57 (m, 2H, CH₂), 2.50 – 2.33 (m, 12H, CH₃

13

+ CH₂). C-NMR (75 MHz, CDCl₃, δ) 174.3 (CO), 143.4 (C_ar), 137.0 (C_ar), 129.9 (CH_ar), 127.4

(CH_ar), 53.2 (CH₂), 52.3 (OCH₃), 48.4 (CH₂), 40.7 (CH₂), 32.2 (CH₂), 21.6 (CH₃). MS (ESI): m/z (%)

= 498.17 (100) [MH]⁺. Anal. Calcd for C₂₂H₃₁N₃O₆S₂: C, 53.10; H, 6.28; N, 8.44. Found: C, 53.25;

H, 6.51; N, 8.53.

6a: ¹H NMR (300 MHz, CDCl₃, δ) 7.72 (d, J = 8.1 Hz, 2H, H_ar), 7.27 (d, J = 8.2 Hz, 2H, H_ar), 5.50

(m, 2H, NH), 3.65 (s, 3H, OCH₃), 2.95 (m, 2H, CH₂), 2.73 (t, J = 6.4 Hz, 2H, CH₂), 2.71 (m, 2H,

CH₂), 2.42 (s, 3H, CH₃) overlapping with 2.41 (m, 2H, CH₂). MS (ESI): m/z (%) = 302.11 (100)

[MH]⁺. Anal. Calcd for C₁₃H₂₀N₂O₄S: C, 51.98; H, 6.71; N, 9.33. Found: C, 51.89; H, 6.62; N, 9.11.

Synthesis of 5a from 6a. A solution of N-tosyl aziridine (278 mg, 1.45 mmol in 5.0 mL of toluene)

was added dropwise to a solution of compound 6a (364 mg, 1.21 mmol in 5.0 mL of toluene) at 60

°C. The mixture was left to react for 8h at reflux temperature. The solvent was evaporated to dryness

and the crude was purified by flash chromatography (SiO₂, AcOEt/n-hexane 7:3 gradient to AcOEt/n-

hexane 9:1) yielding to 5a (503 mg, yield 84%) as a white foam. The spectral data are consistent with

those reported above.

23

RSC Advances

Page 24 of 34

View Article Online

Synthesis of 5b and 6b. N()-Cbz-Lysine methyl ester hydrochloride (0.608 g, 1.8 mmol) was

DOI: 10.1039/C6RA22231E

suspended in CH₃CN (6.0 mL). TEA (0.37 mL, 2.7 mmol) was added, the mixture was warmed at

60°C and half of a solution of N-tosyl aziridine (1.05 g, 5.35 mmol in 16.0 mL of toluene) was added

dropwise in 10 min. After one hour was added dropwise the second half of the solution and the

mixture was left to react for 6h at reflux temperature. The reaction solvent was evaporated to dryness

and the crude was purified by flash chromatography (SiO₂, AcOEt/n-hexane 6:4 gradient to AcOEt/

MeOH 20%) yielding to (6b) (355 mg, yield 40%) and (5b) (552 mg, yield 45%).

5b: ¹H NMR (400 MHz, CDCl₃, δ) 7.75 (d, J = 8.2 Hz, 4H, H_ar), 7.35 – 7.28 (m, 9H, H_ar), 5.49 (d, J

= 8.1 Hz, 1H, NH-Cbz), 5.12 (s, 2H, CH₂), 4.35 (m, 1H, CH), 3.73 (s, 3H, OCH₃), 2.93 (m, 4H, CH₂),

2.51 (m, 4H, CH₂), 2.41 – 2.31 (m, 8H, CH₃+ CH₂), 1.78 (m, 1H, CH₂), 1.64 (m, 1H, CH₂), 1.41 (m,

2H, CH₂), 1.33 (m, 2H, CH₂). ¹³C-NMR (100 MHz, CDCl₃, δ) 173.0 (CO), 158.5 (CO), 143.6 (C_ar),

137.0 (C_ar), 136.4 (C_ar) 129.9 (CH_ar), 128.7 (CH_ar), 128.3 (CH_ar), 128.2 (CH_ar), 127.3 (CH_ar), 67.2

(CH₂), 60.5 (CH₂), 54.1 (CH₂), 53.7 (CH), 52.7 (OCH₃), 40.3 (CH₂), 32.5 (CH₂), 23.0 (CH₂), 21.7

(CH₃). MS (ESI): m/z (%) = 689.26 (100) [MH]⁺. Anal. Calcd for C₃₃H₄₄N₄O₈S₂: C, 57.54; H, 6.44;

N, 8.13. Found: C, 57.10; H, 6.47; N, 8.24. []_D²⁰= 5.60° (c 1.00 in CHCl₃).

6b: ¹H NMR (400 MHz, CDCl₃, δ) 7.76 (d, J = 8.2 Hz, 4H, H_ar), 7.40 – 7.28 (m, 7H, H_ar), 5.43 (d, J

= 7.0 Hz, 1H, NH-Cbz), 5.10 (s, 2H, CH₂), 4.35 (m, 1H, CH), 3.74 (s, 3H, OCH₃), 3.03 (m, 2H, CH₂),

2.76 (m, 2H, CH₂), 2.55 (m, 2H, CH₂), 2.41 (s, 3H, CH₃), 1.78 (m, 1H, CH₂), 1.65 (m, 1H, CH₂), 1.51

(m, 2H, CH₂), 1.35 (m, 2H, CH₂). MS (ESI): m/z (%) = 492.21 (100) [MH]⁺. Anal. Calcd for

C₂₄H₃₃N₃O₆S: C, 58.64; H, 6.77; N, 8.55. Found: C, 58.85; H, 6.74; N, 8.42.

Synthesis of 5b from 6b. A solution of N-tosyl aziridine (168 mg, 0.85 mmol in 4.0 mL of toluene)

was added dropwise to a solution of compound 6b (350 mg, 0.71 mmol in 4.0 mL of toluene) at 60

°C. The mixture was left to react for 8h at reflux temperature. The solvent was evaporated to dryness

and the crude was purified by flash chromatography (SiO₂, AcOEt/n-hexane 7:3 gradient to AcOEt/n-

hexane 9:1) yielding to 5b (442 mg, yield 90 %) as a white foam. The spectral data are consistent

with those reported above.

Synthesis of 7a. Compound 5a (201 mg, 0.40 mmol) was dissolved in anhydrous CH₃CN (12.0 mL).

2,6-Pyridinedimethanol 2,6-dimesylate (120 mg, 0.40 mmol) and K₂CO₃(165 mg, 1.2) were added

and the resulting mixture was left to react for 6h at reflux temperature. The solvent was evaporated

to dryness and the crude was suspended in 20 mL of water and extracted with AcOEt (3 x 10.0 mL).

The organic phase was washed with brine (2 x 15.0 mL), treated with Na₂SO₄and filtered. The solvent

was evaporated to dryness yielding to 7a (232 mg, yield 97%) as a light yellow solid. ¹H NMR (400

MHz, CDCl₃, δ) 7.77 – 7.69 (m, 5H, H_ar), 7.37 – 7.29 (m, 6H, H_ar), 4.32 (s, 4H, CH₂), 3.64 (s, 3H,

24

Page 25 of 34

RSC Advances

_{View A}1_rt3_{icle Online}

OCH₃), 3.10 (m, 4H, CH₂), 2.65 (m, 2H, CH₂), 2.44 (s, 6H, CH₃), 2.35 – 2.24 (m, 6H, CH₂). C-

DOI: 10.1039/C6RA22231E

NMR (100 MHz, CDCl₃, δ) 172.8 (CO), 155.2 (C_ar), 143.7 (C_ar), 138.9 (CH_ar), 130.0 (CH_ar), 127.3

(CH_ar), 124.2 (CH_ar), 54.7 (CH₂), 51.7 (CH₂), 51.3 (OCH₃), 51.1 (CH₂), 45.2 (CH₂), 34.0 (CH₂), 22.9

(CH₂), 21.7 (CH₃). MS (ESI): m/z (%) = 601.3 (100) [MH]⁺. Anal. Calcd for C₂₉H₃₆N₄O₆S₂: C, 57.98;

H, 6.04; N, 9.33. Found: C, 57.62; H, 5.92; N, 9.12. UV/vis (5.2 10^-5mol L^-1, CH₂Cl₂in 1-cm

cuvettes): _max[nm] = 234 nm. IR (ATR) : 2949 cm^-1(w), 1733 cm^-1(ν_C=O), 1157 cm^-1.

Synthesis of 7b. Compound 5b (280 mg, 0.40 mmol) was dissolved in anhydrous CH₃CN (14.0 mL).

2,6-Pyridinedimethanol 2,6-dimesylate (123 mg, 0.42 mmol) and K₂CO₃(184 mg, 1.3 mmol) were

added and the resulting mixture was left to react for 6h at reflux temperature. The solvent was

evaporated to dryness and the crude was suspended in 25 mL of water and extracted with AcOEt (3

x 15.0 mL). The organic phase was washed with brine (2 x 20.0 mL), treated with Na₂SO₄and filtered.

The solvent was evaporated to dryness yielding to 7b (318 mg, 97%) as a light yellow solid. ¹H NMR

(300 MHz, CDCl₃, δ) 7.73 – 7.69 (m, 5H, H_ar), 7.38 – 7.27 (m, 11H, H_ar), 5.29 (m, 1H, NH-Cbz),

5.11 (s, 2H, CH₂), 4.39 – 4.28 (m, 5H, CH and CH₂), 3.74 (s, 3H, OCH₃), 3.10 (m, 4H, CH₂), 2.44 (s,

6H, CH₃), 2.32 – 2.20 (m, 6H, CH₂), 1.79 (m, 1H, CH₂), 1.63 (m, 1H, CH₂), 1.36 – 1.16 (m, 4H,

CH₂). ¹³C-NMR (75 MHz, CDCl₃, δ) 173.1 (CO), 155.1 (C), 143.7 (C_ar), 138.9 (CH_ar), 136.2 (C_ar),

130.0 (CH_ar), 128.7 (CH_ar), 128.3 (CH_ar), 127.3 (CH_ar), 124.2 (CH_ar), 67.1 (CH₂), 55.2 (CH₂), 54.6

(CH₂), 54.0 (CH₂), 52.5 (OCH₃), 51.8 (CH₂), 44.8 (CH₂), 32.6 (CH₂), 27.78 (CH₂), 22.8 (CH₂), 21.7

(CH₃). MS (ESI): m/z (%) = 792.4 (100) [MH]⁺. Anal. Calcd for C₄₀H₄₉N₅O₈S₂: C, 60.66; H, 6.24;

N, 8.84. Found: C, 60.28; H, 5.96; N, 8.60. []_D²⁰= 3.82° (c 1.00 in CHCl₃). UV/vis (5.0 10^-5mol L^-

¹, CH₂Cl₂in 1-cm cuvettes): _max[nm] = 234 nm. IR (ATR) : 3337 cm^-1(NH), 2932 (w), 1750 cm^-1

(ν_C=O), 1152 cm^-1.

Synthesis of 7c. Under H₂atmosphere, compound 7b (700 mg, 0.88 mmol) was dissolved in MeOH

(15.0 mL). Pd/C (70 mg) was added and the resulting suspension was left to react for 24h at rt. The

suspension was filtered on celite layer and celite was washed with MeOH (3 x 5.0 mL). The solvent

was evaporated to dryness yielding to 7c (579 mg, 99%) as a pale yellow foam. ¹H NMR (300 MHz,

CD₃OD, δ) 7.84 (d, J = 8.3 Hz, 4H, H_ar), 7.75 (pst, J = 7.7 Hz, 1H, H_ar), 7.75 (d, J = 8.3 Hz, 4H, H_ar),

7.26 (d, J = 7.7 Hz, 2H, H_ar), 4.59 (m, 2H, CH₂), 4.32 (m, 2H, CH₂), 4.15 (m, 1H, CH) overlapping

with 4.7 – 4.13 (m, 2H, CH₂), 3.88 (s, 3H, OCH₃), 3.75 (m, 2H, CH₂), 3.70 – 3.48 (m, 4H, CH₂), 3.16

(m, 2H, CH₂), 2.45 (s, 6H, CH₃), 2.14 (m, 2H, CH₂), 1.97 (m, 2H, CH₂), 1.73 – 1.59 (m, 2H, CH₂).

¹³C-NMR (75 MHz, CDCl₃, δ) 171.2 (CO), 159.3 (C_ar), 146.5 (C_ar), 140.3 (CH_ar), 134.7 (C_ar), 131.4

(CH_ar), 129.2 (CH_ar), 122.6 (CH_ar), 53.8 (OCH₃), 53.7 (CH), 53.4 (CH₂), 50.3 (CH₂), 46.7 (CH₂),

25

RSC Advances

Page 26 of 34

+

View Article Online

43.5 (CH₂), 31.1 (CH₂), 23.3 (CH₂), 21.5 (CH₃), 20.5 (CH₂). MS (ESI): m/z (%) = 658._D3_O(_I:1₁0_0.0₁₀)₃[₉M_/C6H_RA]₂₂._231E

Anal. Calcd for C₃₂H₄₃N₅O₆S₂: C, 58.42; H, 6.59; N, 10.65. Found: C, 58.54; H, 6.50; N, 10.71.

Synthesis of 7d. Compound 7c (579 mg, 0.88 mmol) was dissolved in AcOH (1.5 mL). NaCNBH₃

(166 mg, 2.65 mmol) was added and the mixture was left to stir 10 min at rt. Acetaldehyde (1.1 mL,

19.36 mmol) was added in 72 h in small amounts (4 eq every 12h). The solvent was evaporated to

dryness, then brine (10.0 mL) was added and the solution was extracted with DCM (5 x 10.0 mL).

The organic phase was treated with Na₂SO₄and filtered. The solvent was evaporated to dryness

yielding to 7d (376 mg, yield 60%) as a light yellow foam. ¹H NMR (400 MHz, CDCl₃, δ) 7.72 (d, J

= 8.2 Hz, 5H, H_Ar), 7.32 (t, J = 8.5 Hz, 6H, H_Ar), 4.34 (s, 4H, CH₂), 3.67 (s, 3H, OCH₃), 3.31 (t, J =

7.4 Hz, 1H, CH), 3.15 – 3.03 (m, 4H, CH₂), 2.70 (dq, J = 14.5, 7.2 Hz, 2H, CH₂), 2.51 – 2.40 (m, 8H,

CH₂+ CH₃), 2.27 (dd, J = 16.6, 9.9 Hz, 6H, CH₂), 1.60 (ddd, J = 15.6, 14.6, 9.2 Hz, 2H, CH₂), 1.31

(m, 2H, CH₂), 1.02 (t, J = 7.1 Hz, 6H, CH₃). ¹³C NMR (100 MHz, CDCl₃, δ) 174.2 (CO), 155.1 (C_Ar),

143.6 (C_Ar), 138.9 (CH_Ar), 136.2 (C_Ar), 130.0 (CH_Ar), 127.3 (CH_Ar), 124.2 (CH_Ar), 110.7 (CH), 63.2

(CH), 55.5 (CH₂), 54.8 (CH₂), 51.7 (CH₂), 51.1 (CH), 44.7 (CH₂), 29.9 (CH₂), 28.3 (CH₂), 24.1 (CH₂),

21.7 (CH₃), 14.1 (CH₃). MS (ESI): m/z (%) = 712.3 (100) [MH]⁺. Anal. Calcd for C₃₆H₅₁N₅O₆S₂: C,

60.56; H, 7.20; N, 9.81. Found: C, 60.22; H, 7.12; N, 10.02.

Reaction of ligand 7c with (R)-(-)-α-Methoxy-α-(trifluoromethyl)phenylacetyl chloride.

(R)-(-)-α-Methoxy-α-(trifluoromethyl)phenylacetyl chloride (17 mg, 0.067 mmol) was added to a

solution of ligand 7c (40 mg, 0.061 mmol) in CD₃COCD₃(1.0 mL), in the presence of pyridine (14

mg, 0.18 mmol). The solution was stirred for 30 min, and then it was transferred into a NMR test

tube. ¹⁹F NMR (282 MHz, CDCl₃, δ) -71.24.

Reaction of ligand 7c with α-Methoxyphenylacetic acid.

To a solution of ligand 7c (21 mg, 0.032 mmol) in CDCl₃(1.0 mL) were added, in this order, (S)-(+)

or (R)-(-)-α-Methoxyphenylacetic acid (5 mg, 0.032 mmol), N,N'-dicyclohexylcarbodiimide (13 mg,

0.065 mmol) and 4-dimethylaminopyridine (3 mg, 0.022 mmol). The solution was stirred for 30 min,

and then it was transferred into a NMR test tube.

General procedure for the synthesis of Ag(I) complexes.

Under N₂, the ligand (0.39 mmol) was dissolved in anhydrous C₂H₄Cl₂(10.0 mL), Ag(I) was added

(0.39 mmol) as AgBF₄or AgOTf and the mixture was left to react for 1h at rt. The suspension was

filtered under N₂, the solvent was concentrated to ~ 1.0 mL and anhydrous n-hexane (10.0 mL) was

added. The mixture was left for 10 min at rt, and then the solvent was evaporated to dryness. n-hexane

26

Page 27 of 34

RSC Advances

View Article Online

(10.0 mL) was added and the suspension was left under stirring over night at rt, then_Dw_OIa_{: 1}s_0.o₁₀p₃e_9/n_Ce₆d_RA2a₂t_231E

air and filtered yielding to the corresponding Ag(I) complex as a white powder.

8a-OTf. White powder: yield 89%; ¹H NMR (300 MHz, CDCl₃, δ) 7.86 (t, J = 7.7 Hz, 1H, H_Ar), 7.76

(d, J = 8.2 Hz, 4H, H_Ar), 7.46 (d, J = 8.2 Hz, 4H, H_Ar), 7.31 (d, J = 7.7 Hz, 2H, H_Ar), 5.04 (d, J = 14.6

Hz, 2H, CH₂), 3.72 (s, 3H, OCH₃), 3.69 (d, J = 14.8 Hz, 2H, CH₂), 3.46 (m, 2H, CH₂), 3.04 (m, 2H,

CH₂), 2.95 – 2.63 (m, 6H, CH₂), 2.48 (m, 8H, CH₂+ CH₃overlapping), 2.19 (br, 2H, CH₂). ¹³C NMR

(75 MHz, CDCl₃, δ) 153.5 (C), 145.8 (C), 140.9 (CH_Ar), 130.7 (CH_Ar), 128.3 (CH_Ar), 125.5 (CH_Ar),

56.6 (CH₂), 53.4 (CH₂), 52.9 (CH₃), 48.9 (CH₂), 47.0 (CH₂), 31.9 (CH₂), 21.8 (CH₃). One quaternary

carbon was not detected. MS (FAB): m/z (%) = 709 (100), 707 (97); 708 (31); 710 (30) [M – OTf]⁺.

Anal. Calcd for C₃₀H₃₆AgF₃N₄O₉S₃: C, 42.01; H, 4.23; N, 6.53. Found: C, 42.13; H, 4.16; N, 6.49.

UV/vis (5.0 10^-5mol L^-1, CH₂Cl₂in 1-cm cuvettes): _max[nm] = 237, 264 nm. IR (ATR) : 1728 cm^-1

(ν_C=O), 1155 cm^-1.

8a-BF₄. White powder: yield 93%; ¹H NMR (400 MHz, CDCl₃, δ) 7.91 – 7.70 (m, 5H, H_Ar), 7.45 (d,

J = 7.8 Hz, 4H, H_Ar), 7.32 – 7.28 (m, 2H, H_Ar), 5.04 (d, J = 14.5 Hz, 2H, CH₂), 3.79 (s, 3H, OCH₃),

3.72 – 3.60 (m, 2H, CH₂), 3.50 – 3.43 (m, 2H, CH₂), 3.05 (m, 2H, CH₂), 2.92 – 2.76 (m, 4H, CH₂),

2.72 – 2.58 (m, 2H, CH₂), 2.49 (m, 8H, CH₂+ CH₃overlapping), 2.11 (m, 2H, CH₂). ¹³C NMR (100

MHz, CDCl₃, δ) 145.8 (C_Ar), 140.7 (CH_Ar), 130.7 (CH_Ar), 128.3 (CH_Ar), 125.5 (CH_Ar), 56.6 (CH₂),

53.2 (CH₂), 52.8 (CH), 48.9 (CH₂), 46.9 (CH₂), 45.6 (CH₂), 21.8 (CH₃). One quaternary carbon was

not detected. MS (FAB): m/z (%) = 709 (100), 707 (97); 708 (31); 710 (30) [M – BF₄]⁺. Anal. Calcd

for C₂₉H₃₆AgBF₄N₄O₆S₂: C, 43.79; H, 4.56; N, 7.04. Found: C, 43.89; H, 4.42; N, 6.62. UV/vis (5.0

10^-5mol L^-1, CH₂Cl₂in 1-cm cuvettes): _max[nm] = 237, 264 nm. IR (ATR) : 1727 cm^-1(ν_C=O), 1161

cm^-1.

8b-OTf. White powder: yield 77%; ¹H NMR (300 MHz, CDCl₃, δ) 7.85 (t, J = 7.7 Hz, 1H, H_Ar), 7.80

– 7.68 (m, 4H, H_Ar), 7.49 – 7.28 (m, 11H, H_Ar), 5.56 (m, 1H, NH), 5.17 (s, 2H, CH₂), 5.14 – 5.05 (m,

2H, CH₂), 4.52 (m, 1H, CH), 3.76 (s, 3H, OCH₃), 3.73 – 3.67 (m, 2H, CH₂), 3.48 (m, 2H, CH₂), 3.27

(m, 1H, CH₂), 3.03 (m, 1H, CH₂), 2.88 – 2.55 (m, 4H, CH₂), 2.51 (s, 6H, CH₃), 2.43 (m, 2H, CH₂),

2.17 (m, 2H, CH₂overlappling with acetone), 1.95 (m, 1H, CH₂), 1.81 (m, 2H, CH₂), 1.59 (m, 2H,

13

CH₂), 1.25 (m, 2H, CH₂). C NMR (75 MHz, CDCl₃, δ) 145.8 (C_Ar), 140.8 (CH_Ar), 130.6 (CH_Ar),

128.3 (CH_Ar), 125.3 (CH_Ar), 67.9 (CH₂), 56.6 (CH₂), 54.1 (CH₂), 53.7 (CH), 52.7 (OCH₃), 31.1 (CH₂),

30.0 (CH₂), 21.8 (CH₃). ¹³C signals have been attributed by HSQC tecnique. Five quaternary carbons

and three CH₂were not located. MS (FAB): m/z (%) = 900 (100), 898 (94%); 901 (40%); 899 (40%)

[M - OTf]⁺. Anal. Calcd for C₄₁H₄₉AgF₃N₅O₁₁S₃: C, 46.95; H, 4.71; N, 6.68. Found: C, 46.57; H,

4.35; N, 6.18. UV/vis (5.0 10^-5mol L^-1, CH₂Cl₂in 1-cm cuvettes): _max[nm] = 237, 264 nm. IR

(ATR): 1717 cm^-1(ν_C=O), 1156 cm^-1.

27

RSC Advances

Page 28 of 34

1

View Article Online

DOI: 10.1039/C6RA22231E

8b-BF₄. White powder: yield 94%; H NMR (400 MHz, CDCl₃, δ) 7.86 – 7.62 (m, 5H, H_Ar), 7.49 –

7.27 (m, 11H, H_Ar), 5.68 (m, 1H, NH), 5.23 – 4.92 (m, 4H, CH₂), 4.48 (m, 1H, CH), 3.76 – 3.61 (s,

5H, CH₂+ OCH₃), 3.42 (m, 1H, CH₂), 3.27 (m, 1H, CH₂), 2.93 (m, 1H, CH₂), 2.80 – 2.54 (m, 4H,

CH₂), 2.53 – 2.25 (m, 8H, CH₂+ CH₃), 2.06 (m, 1H, CH₂), 1.94 (m, 2H, CH₂), 1.77 (m, 2H, CH₂),

13

1.55 (m, 2H, CH₂), 1.25 (m, 2H, CH₂). C NMR (100 MHz, CDCl₃, δ) 173.1 (CO), 157.3 (C_Ar),

156.6 (C_Ar), 153.7 (C_Ar), 145.8 (C_Ar), 140.6 (CH_Ar), 139.2 (CH_Ar), 136.6 (C_Ar), 130.7 (CH_Ar), 128.8

(CH_Ar), 128.6 (CH_Ar), 128.0 (CH_Ar), 125.2 (CH_Ar), 125.1 (CH_Ar), 121.6 (CH_Ar), 67.5 (CH₂), 56.6

(CH₂), 53.8 (CH), 52.7 (OCH₃), 23.1 (CH₂), 21.8 (CH₃). MS (FAB): m/z (%) = 900 (100), 898 (94);

901 (41); 899 (38) [M – BF₄]⁺. Anal. Calcd for C₄₀H₄₉AgBF₄N₅O₈S₂: C, 48.69; H, 5.01; N, 7.10.

Found: C, 48.54; H, 5.16; N, 7.16. UV/vis (5.0 10^-5mol L^-1, CH₂Cl₂in 1-cm cuvettes): _max[nm] =

237, 264 nm. IR (ATR) : 1718 cm^-1(ν_C=O), 1160 cm^-1.

1

8c-BF₄. Yellowish powder: yield 75%; H NMR (300 MHz, CD₃OD, δ) 7.93 – 7.67 (m, 5H, H_Ar),

7.53 – 7.20 (m, 6H, H_Ar), 4.61 (d, J = 17.4 Hz, 2H, CH₂), 4.44 – 4.14 (m, 2H, CH₂), 4.06 (d, J = 17.1

Hz, 2H, CH₂), 3.81 – 3.53 (m, 4H, CH₂), 3.50 – 3.27 (m, 4H, CH₂), 3.20 – 2.91 (m, 3H, CH₂+ CH),

2.45 (s, 6H, CH₃), 2.23 – 1.77 (m, 4H, CH₂), 1.74 – 1.51 (m, 2H, CH₂). ¹³C NMR (75 MHz, CD₃OD,

δ) 159.3 (CO), 146.5 (C_Ar), 140.3 (CH_Ar), 134.7 (C_Ar), 131.4 (CH_Ar), 131.0 (CH_Ar), 129.1 (CH_Ar),

128.2 (CH_Ar), 122.6 (CH_Ar), 67.8 (CH₂), 53.9 (CH), 53.7 (OCH₃), 53.4 (CH₂), 46.7 (CH₂), 43.6

(CH₂), 23.2 (CH₂), 21.5 (CH₃), 20.2 (CH₂). MS (FAB): m/z (%) = 766 (100); 764 (97%); 767 (40%);

765 (38%) [M – BF₄]⁺. Anal. Calcd for C₃₂H₄₃AgBF₄N₅O₆S₂: C, 45.08; H, 5.08; N, 8.21. Found: C,

44.96; H, 5.42; N, 7.93.

8d-BF₄. White powder: yield 83%; ¹H NMR (600 MHz, CDCl₃, δ) 7.78 (br, 5H, H_Ar), 7.46 (br, 4H,

H_Ar), 7.32 – 7.26 (m, 3H, H_Ar), 5.09 (d, J = 13.8 Hz, 2H, CH₂), 3.82 – 3.58 (m, 6H, CH₂), 3.20 (s, 1H,

CH), 3.09 – 2.79 (m, 4H, CH₂), 2.70 – 2.54 (m, 6H, CH₂), 2.48 (s, 6H, CH₃), 2.23 – 2.05 (m, 2H,

CH₂), 2.02 – 1.81 (m, 3H, CH₂), 1.75 – 1.45 (m, 2H, CH₂), 1.27 (br, 2H, CH₂), 1.15 (br, 6H, CH₃).

¹³C NMR (151 MHz, CDCl₃, δ) 153.7 (C_Ar), 145.8 (C_Ar), 140.6 (CH_Ar), 130.7 (CH_Ar), 128.7 (CH_Ar),

128.6 (CH_Ar), 125.0 (CH_Ar), 56.5 (CH₂), 21.8 (CH₃). MS (FAB): m/z (%) = 822 (100); 820 (97%);

821 (40%); 823 (38%) [M – BF₄]⁺; 714 (15%) [MH – AgBF₄]⁺. Anal. Calcd for C₃₆H₅₁AgBF₄N₅O₆S₂:

C, 47.59; H, 5.66; N, 7.71. Found: C, 47.62; H, 5.32; N, 7.75.

General procedure for the Henry reaction

The reactions were performed under a nitrogen atmosphere. The catalyst (0.032 mmoles) was

dissolved in dry CH₂Cl₂(5.0 mL) at rt in a Schlenk flask equipped with a stirring bar (were indicated

molecular sieves-4 Å were added, see Tables1-3 captions). Reactants were then added in the

following order: the proper aldehyde 3 (0.32 mmoles), the nitroalkane (1.6 mmoles), and finally the

selected base (0.032 mmoles). The resulting mixture was stirred at 30 °C for 12h, extracted three

28

Page 29 of 34

RSC Advances

View Article Online

times with brine. The organic layer was dried over sodium sulphate, and the solvent was evaporated

DOI: 10.1039/C6RA22231E

under reduced pressure. The reaction crude was purified by flash column chromatography over a

silica gel column with gradients of n-hexane/ethyl acetate as eluent. Products 4h, 4n and 4p (yields

below 15%) were not isolated.

1

1-(3,5-bis(trifluoromethyl)phenyl)-2-nitroethanol 4j. Yield 66% (64 mg); H NMR (300 MHz,

CDCl₃, δ) 7.94 – 7.89 (m, 3H, H_Ar), 5.64 (pst, J = 6.4 Hz, 1H, CH), 4.74 – 4.48 (m, 2H, CH₂), 3.32 –

13

2

3.06 (bs, 1H, OH). C NMR (75 MHz, CDCl₃, δ) 140.7 (C_Ar), 132.70 (q, J_C,F= 33.8 Hz), 126.4

(CH_Ar), 123.12 (q, ¹J_C,F= 272.8 Hz), 123.05 (hept, ³J_C,F= 3.8 Hz, CH), 80.7 (CH₂), 69.8 (CH). Anal.

Calcd for C₁₀H₇F₆NO₃: C, 39.62; H, 2.33; N, 4.62. Found: C, 39.99; H, 2.60; N, 4.39.

General Procedure for the synthesis of 2-alkynylbenzaldehydes 10a-e and 2-

alkynylnicotinaldehydes 10f-g.

To a solution of 2-bromobenzaldehyde or 2-bromonicotinaldehyde (3.24 mmol) in dry triethylamine

(97.2

mmol),

the

appropriate

alkyne

(3.89

mmol)

and

trans-

dichlorobis(triphenylphosphine)palladium(II) (2 mol%) were added, under a nitrogen atmosphere.

The reaction was stirred at rt for 10 min, and then CuI (1 mol%) was added. The reaction mixture

was stirred at 50 °C until no more starting product was detectable by TLC analysis (eluent: n-

hexane/ethyl acetate). The solvent was then evaporated under reduced pressure and the crude material

was purified by flash chromatography over a silica gel column with gradients of n-hexane/ethyl

acetate as eluent. 2-Alkynylbenzaldehydes 10 a,⁶¹b,⁶¹c,⁸¹d,⁶¹and e,⁶¹and 2-alkynylnicotinaldehydes

f⁸²and g⁸³are known compounds. They were characterised by ¹H-NMR and spectral data are in good

agreement with literature values.

General procedure for the reaction of 2-alkynylarylalehydes 10a-g with nitroalkanes.

The reaction were performed in a 0.25 mmol scale under a nitrogen atmosphere. The catalyst 1-BF₄

(0.025 mmol) was dissolved in dry CH₂Cl₂(1.25 mL) in a Schlenk test tube equipped with a stirring

bar. The mixture was stirred at rt for 10 min. The proper 2-alkynylarylalehyde 10 (0.25 mmol),

nitroalkane (12.5 mmol) and triethylamine (0.025 mmol) were added to the stirred solution, in this

order. In some cases (see Table 5), 50 mg of 4 Å molecular sieves were added. The mixture was

stirred at rt for 1-22 h (for reaction times see Table 5). The reaction mixture was diluted with CH₂Cl₂

(20.0 mL) and the organic layer was washed twice with brine (2  20.0 mL). The organic layer was

dried over sodium sulphate, and the solvent was evaporated under reduced pressure. The reaction

crude was purified by flash column chromatography over a silica gel column with gradients of n-

hexane/ethyl acetate as eluent.

29

Article Doi

Silver comes into play: Henry reaction and domino cycloisomerisation sequence catalysed by [Ag(i)(Pc-L)] complexes

DOI: 10.1039/c6ra22231e

Source and publish data:

Authors:

Article abstract of DOI:10.1039/c6ra22231e

Full text of DOI:10.1039/c6ra22231e

Products guided by the article

R&D Labs maybe for 1029580-46-8

Relevant to this article

Hot Product