Catalytic Activity of Cationic and Neutral Silver(I)-XPhos Complexes with Nitrogen Ligands or Tolylsulfonate for Mannich and Aza-Diels-Alder Coupling Reactions

DOI: 10.1002/chem.201503386

Full Paper

&

Homogeneous Catalysis

Catalytic Activity of Cationic and Neutral Silver(I)–XPhos

Complexes with Nitrogen Ligands or Tolylsulfonate for Mannich

and Aza-Diels–A lder Coupling Reactions

[a]

[b]

[a]

Abdessamad Grirrane,* Eleuterio lvarez, Hermenegildo García,* and Avelino Corma*

Abstract: Cationic and neutral silver(I)–L complexes (L=

Buchwald-type biaryl phosphanes) with nitrogen co-ligands

or organosulfonate counter ions have been synthesised and

characterised through their structural and spectroscopic

properties. At room temperature, both cationic and neutral

silver(I)–L complexes are extremely active catalysts in the

dehyde are efficiently catalysed by these cationic or neutral

silver(I)–L complexes. The solvent influences the catalytic

performance due to limited complex solubility or solvent de-

composition and reactivity. The isolation of new silver(I)–L

complexes with reagents as ligands lends support to mech-

anistic proposals for such catalytic processes. The activity,

stability and metal–distal arene interaction of these silver(I)–

L catalysts have been compared with those of analogous

cationic gold(I) and copper(I) complexes.

3

promotion of the single and double A coupling of terminal

(

di)alkynes, pyrrolidine and formaldehyde. In addition, the

aza-Diels–Alder two- and three-component coupling reac-

tions of Danishefsky’s diene with an imine or amine and al-

Introduction

have been tested as catalysts in single and double three-com-

3

ponent (A ) Mannich-type coupling reactions, providing high-

[

11]

Compared with silver, gold and copper are considered among

the best transition metals for activating and promoting a var-

iety of organic reactions. However, silver(I) has recently also

value propargylamines in high yields and selectivity, often in

minutes, at room temperature. The catalytic activity of these

I

Ag –L complexes has also been extended to two- and three-

[1]

emerged as an efficient metal catalyst. The interest in silver(I)

in catalytic systems stems from its possible coordination

component aza-Diels–Alder coupling reactions of Danishefsky’s

diene with imine or amine and aldehyde, exhibiting good per-

formance and efficiency, and thereby providing a powerful

[

2]

modes and affinity for hard donor atoms such as nitrogen or

[

3]

oxygen. The unstable, poorly active and easily decomposed

methodology for the construction of the nitrogen-containing

[4]

[5]

[6]

[7]

[6,12]

AgI, Ag CO , AgSbF₆and AgOTf salts are used as Lewis

six-membered-ring dihydropyridin-4-one.

Also, the neutral

2

3

acids and as sacrificial agents in some procedures for the prep-

silver(I) complexes [(L1)AgX] and [(L2)AgX] with mixed ligand

combinations L1 or L2 and p-tolylsulfonate (X=p-MeC H SO ),

[8]

aration of gold(I) complexes (Scheme 2, top). In this context,

6

4

3

I

it is important to establish whether or not stable Ag com-

with L being very soft and X being hard, have been seen to be

I

plexes can exhibit good catalytic activity compared with their

catalytically active. The Ag–distal C_a_r_e_n_einteraction in Ag com-

[9]

silver salt precursors in certain reactions. In this work, bulky

cationic 2-(di-tert-butylphosphanyl)biphenyl (L1) and 2-(di-tert-

butylphosphanyl)-2’,4’,6’-triisopropyl)biphenyl (L2; Buchwald

plexes has been analysed by single-crystal X-ray crystallog-

raphy and compared with the corresponding interaction in

analogous Cu– and Au–C_a_r_e_n_ecomplexes. The activity data

show the possible ways in which solvent can influence the cat-

alytic performance due to limited complex solubility or solvent

coordination, reaction or decomposition. For instance, in

[

10]

I

phosphanes) Ag complexes have been synthesised, charac-

terised and their catalytic activity compared with that of their

I

Au and Cu analogues. These cationic silver(I)–L complexes

I

a recent contribution in this area, it was shown that Au and

I

[

a] Dr. A. Grirrane, Prof. Dr. H. García, Prof. Dr. A. Corma

Instituto Universitario de Tecnología Química CSIC-UPV

Universidad PolitØcnica de Valencia

Av. De los Naranjos s/n, 46022 Valencia (Spain)

E-mail: grirrane@itq.upv.es

Cu catalysts in the presence of pyrrolidine react with CH Cl

2

used as reaction solvent to give less-active chloride forms of

[13]

the catalyst.

To expand the scope of these studies, in this work several

I

hgarcia@qim.upv.es

Ag complexes considered as reaction intermediates have been

acorma@itq.upv.es

isolated and characterised, providing an insight into the reac-

tion mechanisms of propargylamine formation and the aza-

Diels–Alder cycloaddition of Danishefsky’s diene. Also, with the

aid of single-crystal crystallography, the lengths of selected

bonds, AgÀN, AgÀP and AgÀCipso^,^aⁿ^d^t^h^e^s^h^o^r^t^e^s^t^dⁱ^s^t^aⁿ^c^e^b^e^-

tween the silver atom and a carbon of the distal phenyl ring in

[

b] Dr. E. lvarez

Instituto de Investigaciones Químicas IIQ-CSIC-US

Consejo Superior de Investigaciones Científicas - Universidad de Sevilla

Av. AmØrico, Vespucio 49, 41092 Sevilla (Spain)

Supporting information and ORCID(s) from the author(s) for this article are

available on the WWW under http://dx.doi.org/10.1002/chem.201503386.

Chem. Eur. J. 2016, 22, 340 – 354

340

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

the synthesised cationic and neutral silver(I)–L complexes have

been measured and compared with those of analogous cop-

per(I) and gold(I) complexes. These data have allowed some

important conclusions to be drawn with respect to the stability

of these complexes and the lability of the aquo, nitrogen and

organosulfonate ligands.

indicated in Scheme 1, the initial phosphine ligand 2 was com-

pletely converted into a mixture of 4 (78%) and 5 (22%). Thus,

the H NMR spectrum shows new signals at d=1.28 and

1

1.23 ppm for the methyl groups of the cationic aquo silver(I)

complex 4 and at d=1.44 and 1.38 ppm for the phosphonium

salt 5 (see Figure S1 in the Supporting Information) instead of

those corresponding to the initial phosphine ligand 2 at d=

1

.11 and 1.07 ppm. Similarly to the H NMR spectrum, the

P NMR spectrum also shows the disappearance of the signal

3

1

Results and Discussion

corresponding to the initial phosphine ligand 2 at d=

In the initial stage of our study and with the aim of isolating

1

4

8.11 ppm (singlet) and the appearance of new signals at d=

5.21 ppm (two doublets) for 4 and at d=33.16 ppm for the

+

À

the presumed silver(I)–acetylide complexes, [Ag] [SbF ] (1),

6

bulky 2-(di-tert-butylphosphanyl)biphenyl (2) and phenylacetyl-

ene (3) as ligand and reagent were dissolved in non-dried

CH Cl at room temperature for 24 h in the stoichiometric pro-

phosphonium salt 5 (singlet; see Figure S2 in the Supporting

Information). Complex 4 can be obtained pure by the reaction

2

+

À

of [Ag] [SbF ] (1) and the phosphine ligand 2 in a stoichio-

6

portions 1:1:2 (Scheme 1, top). Filtration of the resulting trans-

parent, colourless solution followed by addition of n-hexane at

À88C led to the formation of a mixture of a white precipitate

and colourless needle crystals (see the Experimental Section)

that could be isolated by filtration.

metric ratio of 1:1 in non-dried CH Cl at room temperature for

2

8

h (Scheme 1, middle), followed by crystallisation in a mixture

of non-dried dichloromethane/n-hexane (1:2) at À308C for

4 h. After filtration, complex 4 was isolated in a yield of 84%

see the Experimental Section) and characterised by analytical

2

(

1

31

H and P NMR spectroscopy of this solid mixture in CD Cl₂

2

and spectroscopic data (see Table S1 and Figures S3–S6 in the

Experimental Section). The ESI mass spectrum of a solution ob-

tained by dissolving complex 4 in CH Cl /MeOH (1:1) shows

provided evidence of the presence of two kinds of coordinated

phosphine ligand (see the Experimental Section and Figures

S1and S2 in the Supporting Information). Under the conditions

2

a weak positive MS peak at 463.1 Da and an intense positive

MS peak at 405.1 Da (see Figure S5 in the Supporting Informa-

tion), which are attributable, respectively, to the cationic

À

+

[

C H AgF OPSb

(4)ÀSbF +K]

and

[C H AgF OPSb

20 28 6

20

28

6

À

+

(

4)ÀSbF ÀH O] species, and intense negative MS peaks at

6

2

À

2

34.6 and 236.6 Da corresponding to the [SbF ] counter anion

6

(

see Figure S5 in the Supporting Information). The formation of

a silverÀphosphorus bond in complex 4 was firmly confirmed

by the appearance in the P NMR spectrum of two doublets

31

107

due to the coupling of P with the two silver isotopes ( Ag

109

1

107

31

and

14.28 Hz and J( Ag- P)=824.42 Hz (see Figure S4 in the

Supporting Information). On the other hand, in our previous

Ag) of spin = at d=46.08 ppm with J( Ag- P)=

2

1

109

31

7

[

14]

work on copper complexes, we noted that a phosphonium

salt with hexafluorophosphate as the counter anion instead of

hexafluoroantimonate, as in the case of 5, presents a similar

3

1

P chemical shift at d=33.11 ppm (see Figure S2 in the Sup-

porting Information for the spectrum of 5 and our previous

[

14]

work for that of a similar compound with copper). The com-

bustion elemental analysis data for complex 4 is in accordance

with the percentages expected for its formula (see the Experi-

mental Section).

I

The solid-state structure of cationic aquo–Ag –L complex 4

was solved by single-crystal X-ray diffraction studies and the

corresponding ORTEP drawing of complex 4 is shown in

Scheme 1 (bottom, for more details, see Table S1 and Figure S6

in the Supporting Information). The bond lengths of AgÀO,

Scheme 1. Top: Synthesis of cationic aquo–silver(I) complex 4 by simultan-

eous addition of reagents 1, 2 and 3 in a 1:1:2 stoichiometry to non-dried

AgÀP and AgÀC7 , and the shortest distance between the

ipso

silver atom and the carbon (C12) of the distal phenyl ring were

determined to be 2.148(5), 2.3651(15), 2.976 and 2.885 , re-

spectively. The structure obtained is a polymorph of a previous-

CH

2

Cl

2

at room temperature. Phosphonium salt 5 was also detected as prod-

by 2. Middle: Synthesis

uct derived from the formal neutralisation of HSbF

6

of pure cationic aquo–silver(I) complex 4. Bottom: ORTEP view of the struc-

ture of 4. Ellipsoids are drawn at the 30% probability level (CÀH hydrogen

atoms have been omitted for clarity). Single-crystal X-ray data, ORTEP and

crystal-packing details are given in Table S1 and Figure S6 in the Supporting

Information.

[3]

ly reported structure for complex 4.

Another additional unsuccessful experiment aimed at isolat-

ing the silver(I)–acetylide complex 6 (Scheme 2, top) was per-

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

341

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Scheme 2. Top: Failure to isolate any silver–acetylide complex. Middle: Syn-

thesis of cationic complex 7. Bottom: ORTEP view of the structure of 7. El-

lipsoids are drawn at the 30% probability level (CÀH hydrogen atoms have

been omitted for clarity). Single-crystal X-ray data and crystal-packing details

are given in Table S2 and Figure S7 in the Supporting Information.

Scheme 3. Top: Synthesis of cationic silver(I) complexes 10 and 11. Bottom:

ORTEP views of 10 and 11. Ellipsoids are drawn at the 30% probability level

(

CÀH hydrogen atoms have been omitted for clarity). Single-crystal X-ray

formed by starting with the same conditions of the first experi-

ment (Scheme 1), but by using in this case dried CH Cl . The

data and crystal-packing details are provided in Tables S3 and S4 and Fig-

ures S10 and S16 in the Supporting Information, respectively.

2

addition of n-hexane to the resulting transparent colourless so-

lution, followed by standing overnight at À308C did not lead

to the isolation of any solid product. When toluene was added

to this mixture, and after slow crystallisation at room tempera-

ture, the arene–silver(I) complex 7 (Scheme 2, middle) was iso-

lated in a yield of 88% (see the Experimental Section). The

structure of 7 (Scheme 2, bottom) was confirmed by single-

crystal X-ray diffraction (see Table S2 and Figure S7 in the Sup-

At this stage of the work, and with the aim of determining

I

the activity of Ag–L complexes as catalysts, a new cationic

I

Ag –dialkylbiarylphosphane complex was synthesised that can

be used as a pre-catalyst. A stoichiometric (1:1:1) mixture of 1,

2 and aniline (8) or pyrrolidine (9) was stirred for 18 h at room

temperature in CH Cl or toluene (Scheme 3, top). Then the re-

2

sulting mixture was diluted with an additional 2 mL of CH Cl ,

2

porting Information). The distances for AgÀh -toluene (C21

filtered and the supernatant layered carefully with 2 mL of n-

pentane. After standing for 48 h at À308C, colourless crystals

of complex 10 were isolated in a yield of 89% (see the Experi-

mental Section). For the reaction in toluene with pyrrolidine

(9) as nucleophile, the precipitate initially formed was dis-

solved by adding 2 mL of 1,2-dichloroethane, filtered and then

a layer of n-pentane was carefully added, which resulted in the

isolation of colourless crystals of complex 11 in a yield of 91%

and C22), AgÀP and AgÀ(C7) and the shortest distance be-

ipso

tween the silver atom and the closest carbon atom (C8) of the

distal phenyl ring are 2.367(6) and 2.480(6), 2.3873(14), 2.906

and 2.838 , respectively. The structure of 7 is coincident with

[3]

2

the previous reported structure for this complex 7 with h -

I

arene coordination of toluene to Ag. Complex 7 was also char-

acterised by NMR spectroscopy and combustion elemental

analysis (see the Experimental Section and Figures S8 and S9 in

the Supporting Information).

I

(see the Experimental Section). These new cationic Ag–aniline

(10) and –pyrrolidine (11) complexes were fully characterised

by combustion elemental analysis and spectroscopic data (see

the Experimental Section as well as Figures S10–S15 and

Table S3 for 10 and Figures S16–S20 and Table S4 for 11 in the

Supporting Information).

From the experiments previously commented above, it can

be concluded that silver(I) does not form s,p-disilver or any

I

other type of isolable phenylacetylene adduct. Therefore, Ag

I

behaves similarly to Cu , for which it was also not possible to

[13,14]

isolate adducts with phenylacetylene,

and exhibits differ-

Single crystals of complexes 10 and 11 were air-stable and

of suitable quality for an X-ray crystallographic diffraction

study. The solid-state structures of the new cationic silver(I)

I

ent chemistry to Au , which forms fluxional dicationic

[15]

[

(s,p)(LAu) (m-phenylacetylene)][SbF ] complexes.

2 6 2

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

342

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

I

109

complexes 10 (L1–aniline–Ag ) and 11 (L1–pyrrolidine–Ag ) are

Ag isotopes (see Figure S15 in the Supporting Information).

closely related to the previously reported hexafluorophosphate

In negative MS mode, the presence of peaks at 234.8 and

I

I [13]

À

salts of Cu analogues (L1–aniline–Cu) and (L1–pyrrolidine–

236.9 Da, attributable to the SbF counter anion of complex

6

I

[13,14]

121

123

Cu),

respectively, but with a different counter ion. The

10 with Sb and Sb isotopes, can also be observed (see Fig-

ure S15 in the Supporting Information).

bond lengths of AgÀN, AgÀP and AgÀC7_i_p_s_oand the distance

between the silver atom and the closest carbon atom (C12) of

the distal phenyl ring in complex 10 (see Table S3 and Fig-

ure S10 in the Supporting Information) are 2.189(11), 2.374(3),

With the aim of gaining an understanding of the stability

I

and activity of Ag coordinated to bulky ligand 2 in complexes

3

7, 10 and 11, the Mannich A coupling reaction of phenyl-

2

.902 and 2.954 , respectively. Also, the bonds lengths of AgÀ acetylene (3), pyrrolidine (9) and aqueous formaldehyde (15)

N1A, AgÀP and AgÀC7 and the distance between the silver

was performed in toluene by using as catalysts the cationic sil-

ver(I) complexes 7, 10 and 11 (see Table 1). For the sake of

comparison, the catalytic activity of related copper and gold

ipso

atom and the closest carbon atom (C8) of the distal phenyl

ring in complex 11 (see Table S4 and Figure S16 in the Sup-

porting Information) are 2.182(10), 2.3697(13), 3.090 and

[

13]

complexes [LCu(aniline)][SbF ] (12),

[LCu(pyrrolidine)][SbF₆]

6

[

13,14]

[15b]

2

.812 , respectively, which are longer than the corresponding

(13)

and [LAu(aniline)][SbF ] (14)

were also screened

6

values for the CuÀN, CuÀP and CuÀC_i_p_s_obonds and the short-

est distance between Cu1 and the closest carbon atom (C13)

of the distal phenyl ring of the complexed biphenyl in our re-

(see Table 1) with the goal of shedding light on the real silver(I)

active species involved in the catalytic process by comparing

[

13,14]

with the previously studied copper(I)

and gold(I) com-

I

[15]

cently published cationic Cu analogues (L1–aniline–Cu )

plexes. In these catalytic reactions, the expected product, 1-

(3-phenylprop-2-ynyl)pyrrolidine (16), was obtained in good

yields (Table 1, entries 1–6). The product was isolated as

[

13]

(

1.964(2), 2.182(7), 2.786 and 2.577 , respectively) and (L1–

I

pyrrolidine–Cu) (1.956(6), 2.189(19), 2.861 and 2.540 , respect-

[

13]

1

13

ively). It is worth commenting that our previous published

a light-yellow oil and characterised by H and C NMR spec-

troscopy as well as by GC-MS (see Figures S21–S23 in the Sup-

I

[15b]

cationic gold(I) L1–aniline–Au complex

presented inter-

mediate bond lengths compared with the silver and copper

analogues in the case of AuÀN1 and AuÀP1 with lengths of

3

2

.118(8) and 2.250(2) , respectively, and a longer distances in

Table 1. A coupling reaction of phenylacetylene, pyrrolidine and aque-

I

ous formaldehyde with Ag complexes 7, 10, 17, 18, 29 and 27, Cu com-

plexes of 12 and 13 and Au complexes 14 as catalysts

the case of AuÀC13_i_p_s_oand the distance between Au1 and the

I

closest carbon atom (C14) of the distal phenyl ring of around

2

.985 and 3.211 , respectively, thus showing the influence of

the metal (Cu, Ag and Au) on the metal–arene interaction.

The silver(I) complexes 10 and 11 were also characterised by

NMR spectroscopy and ESI-MS (see the Experimental Section

[b]

Entry

Catalyst

T [8C]

t [h]

Conv. [%]

Yield [%]

1

13

31

and the Supporting Information). H, C, DEPT and P NMR

spectra recorded in CD Cl provided evidence that the starting

1

2

3

4

5

6

7

RT

50

0.1

0.2

100

4

90

100

1

100

13

48

52

99

98

99

98

99

trace

81

99

trace

99

10

46

50

97

99

100

2

10

11

12

13

14

AgSbF (1) was completely converted into 10 or 11 depending

6

on the amine (see Figures S11–S14 for 10 and Figures S17–S20

for 11 in the Supporting Information). Thus, the formation of

an AgÀP bond in complexes 10 and 11 was confirmed by the

24

31

appearance in the P NMR spectra of two doublets due to the

7

8

9

10

17

18

RT

0.1

0.05

2

3

1

107

coupling of P with the two silver isotopes of spin = ( Ag

2

109

1

107

31

[c]

11

and Ag) at d=47.35 ppm with J( Ag- P)=639.17 Hz and

[c]

1

109

31

17

29

J( Ag- P)=738.97 Hz for 10 (see Figure S12 in the Support-

ing Information) and at d=45.50 ppm with J( Ag- P)=

11

1

107

31

1

109

31

[d]

6

08.48 Hz and J( Ag- P)=701.65 Hz for 11 (see Figure S18 in

the Supporting Information). Similarly, the H NMR spectra

12

29

27

RT

0.05

1

0

.3

0.05

.1

0.1

99

98

100

[c]

1

3

4

show new signals for the methyl groups of complexes 10 and

0

1

1, respectively, at d=1.20 and 1.15 ppm for 10 (see Fig-

ure S11 in the Supporting Information) and at d=1.29 and

.24 ppm for 11 (see Figure S17 in the Supporting Information)

[c]

1

compared with those corresponding to the initial phosphine

ligand 2 at d=1.11 and 1.07 ppm. The ESI mass spectrum of

a solution obtained by dissolving complex 10 in MeOH shows

a cluster of peaks with major positive MS peaks at 405.1 and

I

[a] All reactions were carried out by using 0.25 mmol of phenylacetylene,

.35 mmol of pyrrolidine, 0.7 mmol of aqueous formaldehyde and 1 mL

4

07.1 Da, attributable to the cationic Ag complex

0

À

+

107

109

[

C H AgF NPSb (10)ÀSbF Àaniline] with

Ag and

Ag

1

26

33

6

of toluene. [b] The conversions and yields were determined by H NMR

spectroscopy and GC of the crude reaction mixtures. [c] CH Cl was used

isotopes, and another cluster of peaks with a major positive

2

I

instead of toluene as solvent. [d] Dioxane was used instead of toluene as

solvent.

MS peak at 437.2 Da, attributable to cationic Ag complex

À

+

107

[

C H AgF NPSb (10)ÀSbF Àaniline+MeOH] with Ag and

26

33

6

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

343

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

porting Information). It should be noted, however, that the re-

action took place in only few minutes (6 min) at room tem-

I

perature (268C) for Ag complexes 7, 10 and 11 (Table 1, entries

1

–3), which compares with the slightly higher reaction

temperature (508C) and longer reaction time (12 min) required

I

for the reaction with Cu complexes 12 and 13 (Table 1, entries

4

and 5). A considerably longer reaction time (24 h) at 508C

I

was required for the reaction with the cationic Au complex 14

(

Table 1, entry 6).

One important conclusion that can be drawn from the re-

I

sults in Table 1 (entries 1–6) is that the cationic Ag complexes

I

are more active than the Cu complexes, and considerably

I

more active than the Au complex. This higher activity of Ag

I

and Cu compared with Au can be explained by the lack of for-

mation of a dicationic s,p-disilver– (Scheme 2, top) or dicop-

[

14]

per–phenylacetylene adduct analogous to the fluxional di-

[

15a]

cationic [(s,p)(LAu) (m-phenylacetylene)][SbF ] .

On the

2

6 2

I

other hand, the higher catalytic activity of the cationic Ag

I

complex compared with the Cu complexes can be explained

I

by the more labile coordination between Ag and the hard

Scheme 4. Top: Synthesis of cationic silver(I) complex [(L1)Ag(propargyl-

amine)][SbF ] (17). Bottom: ORTEP view of 17. Ellipsoids are drawn at the

6

donor nitrogen atom present in the neutral ligands 8, 9 and

I

16 compared with in the analogous Cu complexes. This ration-

3

0% probability level (CÀH hydrogen atoms have been omitted for clarity).

alisation is supported by the single-crystal crystallographic

Single-crystal X-ray data and crystal-packing details are given in Table S5 and

Figure S28 in the Supporting Information.

I

data for the cationic Ag 10 and 11 complexes compared with

I

those of the cationic L–Cu–aniline and L–Cu–pyrrolidine ana-

logues, in which the bond lengths of AgÀN in complexes 10

and 11 (see Table S3 and Figure S7 in the Supporting Informa-

tion) are around 2.189(11) and 2.182(10) , respectively, longer

than the CuÀN bond lengths of 1.964(2) and 1.956(6) in the

cantly longer than the CuÀN bond length of 1.979(2) in the

[

13]

analogous [(L1)Cu(propargylamine)][PF ]

complex. This dif-

6

ference in bond length shows the stronger coordination of the

I

cationic L–Cu–aniline and L–Cu –pyrrolidine complexes, re-

propargylamine ligand with Cu compared with Ag . Interest-

[

13]

spectively.

ingly, the catalytic activity of the cationic silver(I) complex 17

3

To isolate possible reaction intermediates, the A coupling

reaction was performed in the presence of a stoichiometric

in the Mannich A coupling reaction (Table 1, entry 7) was as

I

high as those of the initial cationic Ag complexes 7, 10 and

+

À

amount of [Ag] [SbF ] (1), bulky 2-(di-tert-butylphosphanyl)bi-

11.

6

3

phenyl (L1=2), phenylacetylene (3), pyrrolidine (9) and aque-

ous formaldehyde (15, 37%) at room temperature in toluene

Previously, we found that the nature of the solvent in this A

coupling reaction plays a key role in solubilising the cationic

pre-catalyst and the subsequent intermediates formed during

the course of the catalytic cycle, making possible the regenera-

(

Scheme 4, top). After 20 h, the precipitate formed was dis-

solved by adding 2 mL of dichloroethane/dichloromethane

1:1) and the resulting transparent solution was filtered, fol-

[

13,14]

(

tion of the active form.

To investigate the behaviour of cat-

lowed by slow evaporation at room temperature to afford col-

ionic silver(I) complexes in CH Cl , in which previously cationic

2

ourless crystals of complex 17 in a yield of 90% (see the Ex-

copper(I) and gold(I) complexes became inactive due to the

perimental Section). The [(L1)Ag(propargylamine)][SbF ] com-

plex (17), which corresponds to the Ag catalyst bearing the

formation of the neutral bridged dichloride dicopper(I) [{L1–

6

I

[13]

Cu(m-Cl)}₂]

and chloride gold(I) [L1–AuCl]

complexes

final product propargylamine 16, was isolated and character-

ised by elemental analysis and spectroscopic data (see Fig-

ures S24–S28 and Table S5 in the Supporting Information). The

combustion elemental analysis data of complex 17 is in ac-

cordance with the percentages expected for its formula. Crys-

tals of complex 17 of sufficient quality for X-ray crystallography

were obtained by slow evaporation of a filtered solution of tol-

uene/CH Cl /ClCH CH Cl (2:1:1) of 17 (see the Experimental

(Scheme 5, top), respectively, new experiments for the prepara-

tion of complexes 11 and 17 in CH Cl instead of toluene were

2

performed (Scheme 5, bottom).

Transparent colourless solutions were observed from the be-

ginning to the end of the reaction (20 h) in the preparation of

complexes 11 and 17 in CH Cl (Scheme 5, bottom) due to the

2

higher solubility of these complexes, reagents and products in

CH Cl compared with in toluene. Importantly, in the case of

2

I

Section and Table S5 and Figure S28). The ORTEP structure and

Ag complexes, the presence of chlorine atoms in the medium,

which could be formed by the reaction of CH Cl with pyrrol-

I

crystal-packing details for the Ag complex 17 are given, re-

2

I

spectively, in Scheme 4 (bottom) and Figure S28 in the Sup-

porting Information.

idine (9), apparently does not deactivate the cationic Ag com-

plexes 11 and 17 by direct coordination of chloride to the

metal ion. In another independent experiment followed by

The bond length of AgÀN in complex 17 (see Table S5 and

Figure S28 in the Supporting Information) is 2.209(2) , signifi-

31

P NMR spectroscopy to prepare 17 using CD Cl (Scheme 5,

2

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

344

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Scheme 6. Top: Synthesis of the neutral dichloride disilver(I) complex 18.

Bottom: ORTEP view of 18. Ellipsoids are drawn at the 30% probability level

(

CÀH hydrogen atoms have been omitted for clarity). Single-crystal X-ray

data and crystal-packing details are given in Table S6 and Figure S37 in the

Supporting Information.

for 4 h at room temperature, led to the formation of colourless

crystals of 18 in a yield of 88% (see the Experimental Section).

This new neutral dichloride-bridged disilver(I) complex [{L1–

Scheme 5. Top: Deactivation of cationic copper(I) and gold(I) complexes by

the formation of less active, neutral bridged dichloride dicopper(I) [{L1-Cu(m-

Ag(m-Cl)} ] (18) was fully characterised by its combustion elem-

2

[

13]

Cl)}

paration of complexes 11 and 17 in CH

a small amount of a less active, neutral bridged dichloride disilver(I) [{L1-

2

] and chloride gold(I) [L1-AuCl] complexes, respectively. Bottom: Pre-

ental analysis and spectroscopic data (see the Experimental

Section as well as Figures S33–S37 and Table S6 in the Sup-

porting Information). Crystals of 18 were air-stable and of suit-

able quality for single crystal X-ray crystallographic diffraction

2

Cl as solvent with the formation of

2 2 2

Ag(m-Cl)} ] complex (18). Preparation of 17 starting from 11 in CD Cl as solv-

ent.

(

Scheme 6, bottom). Thus, according to Scheme 6, insoluble

AgCl can become solubilised in CH Cl when phosphines L are

2

bottom), the complete transformation of complex 11 into 17

was observed (yield 90%) accompanied by two additional

minor products. One of these corresponds to new species 18

and remained at the end of the reaction (less than 5%;

Scheme 5, bottom). Complex 18 shows two doublets due to

present in the medium. This finding has to be put in the con-

I

text of the general practice of using Ag salts to presumably

À

I

[16]

remove Cl from Au, which has led to erratic results with

regard to their catalytic activity.

The solid structure of cationic complex 18 is closely related

31

1

I

the coupling of P with the two silver isotopes of spin =

to that of the previously reported Cu analogue [{L1-Cu(m-

2

107

109

1

107

31

[13]

(

Ag and Ag) at d=43.67 ppm with J( Ag- P)=605.60 Hz

Cl)}₂]. The bond lengths of AgÀCl1, AgÀCl1#1, AgÀP1 and

1

109

31

and J( Ag- P)=698.30 Hz. The second small peak (also less

than 5%), which had disappeared by the end of the reaction,

can be attributed to the plausible cationic intermediate com-

plex 19 (see Scheme 7). This intermediate shows two doublets

AgÀC7 , the distances Ag–Ag and Cl—Cl, and the shortest

ipso

distance between the silver atom and the closest carbon atom

(C8) of the distal phenyl ring of L1 (see Table S7 and Figure S37

in the Supporting Information) are 2.4909(7), 2.6296(7),

2.3941(6), 3.109, 3.393, 3.837 and 2.823 , respectively. Interest-

3

1

31

in the P NMR spectrum due to the coupling of P with the

1

107

109

[13]

two silver isotopes of spin

=

2

(

Ag and

Ag) at d=

ingly the dichloride dicopper(I) [{L1–Cu(m-Cl)}₂] analogue ex-

1

107

31

1

109

31

4

5

5.40 ppm with J( Ag- P)=490.55 Hz and J( Ag- P)=

hibits shorter distances, with the bond lengths CuÀCl1, CuÀ

3

1

67.26 Hz (see Figures S29–32 for the P NMR spectra of mix-

Cl1#1, CuÀP1 and CuÀC7 , the distances Cu—Cu and Cl—Cl,

ipso

tures of complexes of 17, 18 and 19 in the Supporting Infor-

mation).

and the shortest distance between the copper atom and the

closest carbon atom (C8) of the distal phenyl ring being

2.2954(4), 2.3572(4), 2.2005(4), 2.906, 3.0408, 3.522 and 2.559 ,

respectively.

To confirm the identity of the neutral chloride–Ag complex

18, which would be the analogue of the dichloride dicopper(I)

1

31

I

[

(

{L1–Cu(m-Cl)}₂] or chloride gold(I) [L1–AuCl] complexes

Scheme 5, top), we proceeded to its alternative synthesis by

H and P NMR analysis of this neutral dimeric Ag complex

provided evidence of the coordination of the initial free phos-

the reaction of AgCl as precursor with bulky 2-(di-tert-butyl-

phosphanyl)biphenyl (L1=2) in a stoichiometry of 1:1.2 in

dried CH Cl at room temperature for 48 h (Scheme 6, top). Fil-

phine ligand 2 (see the Experimental Section and Figures S33-

S34 in the Supporting Information). Thus, the H NMR spec-

1

trum shows new signals at d=1.30 and 1.25 ppm for the

methyl groups of disilver(I) complex 18 (see Figure S33 in the

Supporting Information) instead of the values corresponding

to the initial phosphine ligand 2 at d=1.11 and 1.07 ppm.

2

tration of the resulting transparent colourless solution to

remove the excess insoluble AgCl, followed by slow evapora-

tion of the resulting mixture in dichloromethane after standing

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

345

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

1

31

Similarly to the H NMR spectrum, the P NMR spectrum of 18

also shows the disappearance of the signal corresponding to

the initial phosphine ligand 2 at d=18.11 ppm (singlet) and

the appearance of a new signal (two doublets) due to the

place spontaneously or be promoted by acids. Silver(I)–prop-

argylamine complex 17 will be formed from the plausible key

silver intermediate 19, followed by the release of propargyl-

amine 16 from the coordination sphere of Ag by replacement

with free pyrrolidine present in the reaction medium.

I

31

1

107

coupling of P with the two silver isotopes of spin = ( Ag

2

109

1

107

31

and Ag) at d=43.67 ppm with J( Ag- P)=605.60 Hz and

According to this mechanistic proposal, the first major differ-

1

109

31

I

I[13,14]

I[13]

J( Ag- P)=698.30 Hz for 18 (see Figure S34 in the Support-

ence between the cationic L1–Ag , L1–Cu

and L1–Au

1

3

ing Information). Also, C and DEPT NMR spectroscopy (see

Figures S35 and S36 in the Supporting Information) shows the

presence of signals exclusively attributable to 18.

complexes used as catalysts in this Mannich A coupling reac-

tion is the lack of formation of s,p-disilver– and dicopper–

phenylacetylene adducts, whereas, in contrast, the correspond-

I [15]

The catalytic activity of this neutral dichloride-bridged disil-

ing adduct is formed in the case of L1–Au.

The second

I

ver(I) complex [{L1–Ag(m-Cl)} ] (18) was also tested in the Man-

nich A coupling reaction (Table 1, entry 8); a considerably

major difference between L1–Ag on one hand and L1–Cu and

L1–Au complexes on the other, is the higher stability of the

cationic Ag intermediates with respect to its inactivation in

2

3

I

lower activity was observed in comparison with the cationic

I

Ag pre-catalysts 7, 10, 11 and 17. The use of dichloromethane

as solvent instead of toluene when cationic Ag complexes 11

the presence of chloride by formation of neutral dichloride. In

contrast, copper and gold deactivate severely in the presence

I

and 17 were used as catalysts (Table 1, entries 9 and 10 com-

of chloride by forming dicopper(I) [{L1-Cu(m-Cl)} ] or chloride

2

pared with entries 3 and 7) led to a higher reaction rate in this

gold(I) [L1-AuCl] complexes when CH Cl is used as solvent.

To expand the scope of this catalytic A coupling reaction,

2

3

catalytic A coupling reaction due to the higher solubility of

these cationic silver(I) complexes in this solvent and their negli-

gible deactivation in the presence of chlorine in the medium

the double reaction of 1,4-diethynylbenzene (20), pyrrolidine

and aqueous formaldehyde with cationic Ag complex 11 was

I

compared with its Cu and Au analogues.

performed in toluene, CH Cl₂or dioxane as solvent (see

2

According to these experimental results, a plausible reaction

Table 2). Control reactions in the absence of any catalyst did

not allow the detection of any product. The expected reaction

3

mechanism for the A coupling reaction with [Ag(L1)(toluene)]

3

[

SbF ] (7) or [Ag(L1)(NH Ph)][SbF ] (10) as pre-catalyst in tolu-

products of the mono- and di-A coupling reactions, 1-[3-(4-

6

2

6

ene or CH Cl can be proposed (Scheme 7). The main feature

ethynylphenyl)prop-2-yn-1-yl]pyrrolidine (21) and 1,4-bis(3-

2

(

pyrrolidin-1-yl)prop-1-yn-1-yl)benzene (22; see scheme in

Table 2), respectively, were obtained in good yields and charac-

3

terised. Moreover the double A coupling took place in a short

I

time when using cationic Ag complex 11 as catalyst (Table 2),

the solvent playing a significant role in the catalytic process.

Toluene as solvent (see Table 2, entry 1) was less convenient

than dioxane or CH Cl (see Table 2, entries 2 and 3). The lower

2

reactivity in toluene is due to the limited solubility of the cat-

ionic silver(I) complex 11 used as pre-catalyst as well as the re-

action intermediates.

Bis-propargylamine compound 22 was isolated and charac-

1

13

terised by H, C and DEPT NMR spectroscopy (see Fig-

ures S38–S40 in the Supporting Information). GC-MS of 22

(

C H N ) dissolved in CH Cl shows a peak at 292.19 Da, in

20 24 2 2 2

agreement with the expected molecular formula (see Fig-

ure S41 in the Supporting Information). In addition to the bis-

propargylamine 22 as the final product, the formation of

mono-propargylamine 21 (C H N) was observed as the pri-

15

3

mary product from the mono-A coupling and characterised by

GC-MS analysis, which showed a peak at 209.2 Da, in agree-

ment with the expected molecular formula (see Figure S42 in

the Supporting Information). A plausible mechanism for the

3

Scheme 7. Proposed mechanism for the A coupling reaction leading to

I

propargylamine 16 using as pre-catalyst the cationic Ag complexes 7 or 10

in the presence of toluene or CH Cl as solvent. Complexes 11 and 17 were

2 2

isolated as intermediates. Complex 19 is proposed as an intermediate (see

Figure S32 in the Supporting Information).

3

double A coupling reaction leading to 22 would be similar to

[

13]

that previous reported for analogous cationic copper(I) com-

plexes as catalyst (see Figure S43 in the Supporting Informa-

tion).

I

of this mechanism is the coordination of Ag pre-catalyst 7 or

1

0 with pyrrolidine leading to the isolated intermediate

To show the generality of the use of the this new cationic

I

3

[

Ag(L1)(pyrrolidine)][SbF ] (11) and the presumed p-complex

Ag complex 11 as catalyst, double A coupling of aliphatic 1,6-

heptadiyne (23) was also tested (Table 2) under the optimised

conditions. The expected bis-propargylamine 25 was obtained

as the final product in good yield and with excellent selectivity

6

intermediate 19 (see Figure S32 in the Supporting Information)

formed between silver and phenylacetylene. The condensation

reaction between formaldehyde and pyrrolidine would take

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

346

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

3

Table 2. Double A coupling reaction of 1,4-diethynylbenzene (20) or 1,6-

I

heptadiene (23), pyrrolidine and aqueous formaldehyde with Ag complex

[a]

11 as catalyst (6 mol%).

[

b]

[b]

Entry

Solvent

T [h]

Conv. [%]

Yield [%]

1

2

3

toluene

dioxane

CH Cl

2 2

0.2

0.1

90

100

87

98

4

5

dioxane

CH Cl

0.2

0.1

100

97

98

2

[

0

a] All reactions were carried out by using 0.25 mmol of the bis-alkyne,

.7 mmol of pyrrolidine, 1.4 mmol of aqueous formaldehyde and 1 mL of

solvent. [b] The conversions and yields, respectively, of 20 or 23 and 22

1

or 25 were determined by H NMR spectroscopy and GC of the crude re-

action mixtures.

(

see Table 2, entries 4 and 5). As expected, the mono-prop-

argylamine 24 was also observed as the primary reaction prod-

uct. Bis-propargylamine 25 was isolated as a light-yellow oil

and characterised by H, C and DEPT NMR spectroscopy (see

Scheme 8. Top: Synthesis of neutral silver(I) complexes L1–Ag–(p-tolylsulfon-

ate)(H O) (27) and L2–Ag–(p-tolylsulfonate) (29). Bottom: ORTEP views of 27

2

and 29. Ellipsoids are drawn at the 30% probability level (CÀH hydrogen

atoms have been omitted for clarity). Single-crystal X-ray data and crystal-

packing details are given in Table S7 and Figure S53 for 27 and Table S8 and

Figure S59 for 29 in the Supporting Information.

1

13

Figures S44–S46 in the Supporting Information). GC-MS of 25

(

C H N ) dissolved in CH Cl shows a peak at 258.2 Da, in

17 26 2 2 2

agreement with the expected molecular formula (see Fig-

1

ure S47 in the Supporting Information). The H NMR spectrum

of 25 shows a coupling between the two CH groups across

(Scheme 2, bottom for ORTEP structures and Tables S7 and S8

and Figures S53 and S59 in the Supporting Information for

single-crystal X-ray data and crystal-packing details).

2

the CꢀC bonds at d=3.25 (triplet) and 2.24 ppm (triplet of

4

triplets) with J=2.20 Hz (see the Experimental Section and

Figure S44). On the other hand, mono-propargylamine com-

The neutral silver(I) complexes 27 and 29 were also charac-

terised by combustion elemental analysis, NMR spectroscopy

and ESI-MS (see the Experimental Section and Figures S49–S54

for 27 and Figures S55–S60 for 29 in the Supporting Informa-

pound 24 (C H N; see scheme in Table 2) was characterised

12

17

by GC-MS, which showed a peak at 174.1 Da, in agreement

with the expected molecular formula (see Figure S48 in the

Supporting Information).

1

13

31

tion). H, C, DEPT and P NMR analysis of solutions in CD Cl₂

2

To expand the scope of the cationic silver(I) complexes as

catalysts and with the aim of confirming the lower activity of

the related neutral complexes, the neutral (phosphine)silver(I)

organosulfonate complexes [(L1)AgX] (27) and [(L2)AgX] (29)

were prepared. A stoichiometric (1:1) mixture of silver(I) p-tolyl-

sulfonate (26) and L1 (2) or L2 (28) was stirred for 24 h at

room temperature in CH Cl (Scheme 8, top), followed by the

provided evidence that the starting silver(I) p-tolylsulfonate

(26) was completely converted into 27 or 29 (see Figures S49–

S52 for 27 and Figures S55–S58 for 29 in the Supporting Infor-

mation). Thus, the formation of an AgÀP bond in complexes

31

27 and 29 was confirmed by the appearance in the P NMR

31

spectra of two doublets due to the coupling of P with the

1

107

109

two silver isotopes of spin = ( Ag and Ag), respectively, at

2

31

1

107

1

109

31

subsequent careful addition of a layer of n-hexane over the re-

sulting transparent solutions, and then kept at À308C for 48 h.

This procedure afforded colourless crystals of complexes 27

and 29, each in a yield of 92% (see the Experimental Section).

The structures of these stable neutral silver(I) complexes 27

and 29 were resolved by single-crystal X-ray crystallography

d=43.93 ppm with J( Ag- P)=695.07 Hz and J( Ag- P)=

802.18 Hz for 27 (see Figure S50 in the Supporting Informa-

tion) and at d=42.09 ppm with J( Ag- P)=685.33 Hz and

1

107

31

1

109

31

J( Ag- P)=791.34 Hz for 29 (see Figure S56 in the Support-

ing Information). The coupling constants are smaller in the

case of complex 29 with the bulky 2-(di-tert-butylphosphanyl)-

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

347

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

2

’,4’,6’-triisopropylbiphenyl (L2) compared with the coupling

Table 3. Aza-Diels–Alder cycloaddition of imine 30 and Danishefsky’s

constants of complex 27 with the less bulky 2-(di-tert-butyl-

phosphanyl)biphenyl (L1; Scheme 8). The formation of the

neutral Ag complexes 27 and 29 was also confirmed by ESI-

I

diene (31) promoted by neutral (27, 29) and cationic (10) Ag complexes

[a]

as catalysts (6 mol%).

I

MS (for details see Figure S54 for 27 and Figure S60 for 29 in

the Supporting Information).

I

The solid-state structures of neutral Ag complexes L1–Ag–

(

p-tolylsulfonate)(H O) (27) and L2–Ag–(p-tolylsulfonate) (29)

2

[b]

Entry

Catalyst Solvent/H

2

O (3:1) T [h] Conv. [%]

Yield [%]

were analysed by X-ray diffraction. The presence of one crystal-

lisation CH Cl molecule and an additional water co-ligand was

1

2

3

4

5

6

7

29

27

10

33

CH

2

Cl

2

/none

2

1

0

99

97

94

8

–

94

91

80

5

2

dioxane/none

dioxane/H O

dioxane/H

CH Cl /H

dioxane/H

I

revealed in the case of L1–Ag –(H O)X (27) with the less bulky

2

phosphine ligand L1 with the water molecule in the trans posi-

tion with respect to the phosphorus atom and an P1ÀAg1À

O4(water) angle of 146.71(5)8 (Scheme 8, middle). In the case

2

[c]

2

O

2

O

2

O

99

96

I

of L2–Ag –X (29) with the more bulky phosphine ligand L2, no

water co-ligand exists and the p-tolylsulfonate (X) is in the

trans position with the phosphorus atom with an angle P1À

Ag1ÀO1(X) of 164.42(6)8 (Scheme 8, bottom). The bond

[

d]

[e]

8

9

1

29

10

33

29

27

dioxane/H O

2

1

99

98

99

98

50

45

86

91

93

92

[

lengths for AgÀO, AgÀP and AgÀC7 and the distance be-

dioxane/H

2

O

ipso

[

f]

[d]

0

tween the silver atom and the closest carbon atom (C8) of the

distal phenyl ring for complex 27 (see Table S7 and Figure S53

in the Supporting Information) are 2.2652(18) (water),

1

[f]

f]

1

13

2

[

2

dioxane/H O

2

.4722(18) (p-tolylsulfonate), 2.3798(6), 2.963 and 2.852(3) , re-

[a] All reactions were carried out by using 0.25 mmol of imine, 0.3 mmol

^s^p^e^c^tⁱ^v^e^l^y^,^w^h^e^r^e^a^s^t^h^e^A^g^À^O^,^A^g^À^P^aⁿ^d^A^g^À^C⁷ipso bonds

lengths and the distance between the silver atom and the clos-

est carbon atom (C8) of the distal phenyl ring for complex 29

of Danishefsky’s diene (31) and solvent (1.5 mL)/H O (0.5 mL). [b] The

conversion of 30 and yield of 32 were determined by H NMR spectros-

2

1

copy and GC of the crude reaction mixture. [c] 7% of aniline and 5% of

benzaldehyde were also observed. [d] The conversion of 8 was deter-

mined by GC of the reaction mixture. [e] About 20% of an unknown

product with a molecular peak at 161 Da was also observed (see Fig-

ure S57 for the GC-MS data). [f] Slow addition of Danishefsky’s diene (31)

to the mixture of aniline (8) and benzaldehyde (35) over 50 min.

(

see Table S8 and Figure S59 in the Supporting Information)

are 2.140(2), 2.3637(7), 3.029 and 2.834(3) , respectively.

The catalytic activities of these new neutral silver(I) com-

3

plexes 27 and 29 for the single Mannich A coupling reaction

were studied under the same conditions as used for complex

11, but by using a different solvent; it was again confirmed

that the choice of solvent was crucial for the success of the

process. When dioxane or CH Cl was used instead of toluene

In the catalytic reaction, neutral complex L2–Ag–(p-tolylsul-

fonate) (29) was found to be inactive as catalyst in CH Cl or

2

as solvent, neutral complexes 27 and 29 became very active,

promoting the formation of propargylamine 16 in very good

yields and selectivities (see Table 1, entries 11–14). This high ac-

tivity in dioxane and CH Cl compared with in toluene can be

dioxane as solvent (Table 3, entries 1 and 2), whereas in the

presence of a mixture of dioxane/water (3:1), the reaction took

place in a short time (1 h) to give the corresponding 1,2-di-

[

6,12]

phenyl-2,3-dihydro-4-pyridone (32)

in high yield and select-

2

I

explained by their greater ability to dissolve the Ag complexes

7 and 29, and also by the principles of hard–soft acid–base

ivity (Table 3, entry 3). Compound 32 was characterised by

NMR spectroscopy and GC-MS (see Figures S61–S64 in the

Supporting Information). Similarly, by using neutral L1–Ag–(p-

2

theory, by which polar solvents should make easier the dissoci-

ation of sulfonate complexes L–Ag–X to afford the cationic

tolylsulfonate)(H O) (27) as catalyst under the same optimised

2

+

[

LAg] complex, which is probably the catalytically active spe-

conditions, a high activity was also observed (Table 3, entry 4).

I

cies.

It has been reported that the aza-Diels–Alder cycloaddition

For the sake of comparison, the catalytic activity of cationic Ag

I

complex 10 (L1–aniline–Ag ) was also evaluated under these

of Danishefsky’s diene with imines proceeds efficiently in water

in the presence of a sub-stoichiometric amount of silver(I) ben-

optimised conditions; pyridone 32 was formed although in

a slightly lower yield, accompanied by small amounts of aniline

and benzaldehyde as by-products as a result of the hydrolysis

of the starting imine 30 (Table 3, entry 5). When CH Cl was

used as solvent instead of dioxane under the same conditions,

a very low conversion and yield were achieved (Table 3,

[6]

zenesulfonate (10%). Aimed at developing a robust catalyst

for this transformation and to shed some light on the mechan-

ism of this catalytic reaction by the isolation of possible inter-

mediates, the catalytic activities of neutral L1–Ag–(p-

2

tolylsulfonate)(H O) (27) and L2–Ag–(p-tolylsulfonate) (29;

entry 6). It was also observed that the mixture THF/H O was

2

Table 1) were investigated. It was observed that complexes 27

and 29 even in lesser amounts (6%) were considerably more

active than silver benzenesulfonate. This higher activity can be

attributed to the higher stability of complexes 27 and 29 pro-

vided by the bulky phosphine ligands.

not a suitable medium because additional isomeric by-prod-

ucts (GC-MS: 130 Da) were formed by the reaction of THF with

Danishefsky’s diene (31).

When the reagents of this aza-Diels–Alder cycloaddition,

imine 30 and Danishefsky’s diene (31), were mixed with a stoi-

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

348

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

[

(L1)Ag(propargylamine)][SbF ] (17), which indicates that imine

6

ligand 30 is labile, allowing operation of turnovers through the

intermediary of the complex [(L1)Ag(imine)][SbF ] (33). The

6

I

ORTEP view of the Ag complex 33 is shown in Scheme 9

(

bottom; see the Experimental Section and Table S9 and Fig-

ure S69 in the Supporting Information).

The new cationic silver(I) complex 33 was characterised by

NMR spectroscopy and ESI-MS (see the Experimental Section

and Figures S65–S68 and S70 in the Supporting Information).

1

13

31

H, C, DEPT and P NMR analysis of solutions in CD Cl pro-

2

vided evidence that the starting AgSbF (1) was completely

6

converted into 33 by coordination of both the starting free lig-

ands phosphine 2 and imine 30 (see Figures S65–S68 in the

Supporting Information). The formation of an AgÀP bond in

31

complex 33 was confirmed by the appearance in the P NMR

31

spectrum of two doublets due to the coupling of P with the

1

107

109

two silver isotopes of spin

=

2

(

Ag and

Ag) at d=

1

107

31

1

109

31

4

7

6.03 ppm with J( Ag- P)=634.08 Hz and J( Ag- P)=

32.02 Hz (see Figure S66 in the Supporting Information). Simi-

1

larly, the H NMR spectrum shows new signals for the methyl

groups of complex 33 at d=1.29 and 1.24 ppm (see Fig-

ure S65 in the Supporting Information) instead of those corres-

ponding to the initial phosphine ligand 2 that appear at d=

1

.11 and 1.07 ppm. The ESI mass spectrum of a solution ob-

Scheme 9. Top: Synthesis of cationic silver(I) complex [(L1)Ag(imine)][SbF

6

]

tained by dissolving complex 33 in CH Cl /MeOH (1:1) shows

2

(

33). Middle: Failure to isolate silver complex 34. Bottom: ORTEP view of 33.

a cluster of ions with major positive peaks at 586.2 and

Ellipsoids are drawn at the 30% probability level (CÀH hydrogen atoms have

been omitted for clarity). Single-crystal X-ray data and crystal packing details

are given in Table S9 and Figure S69 in the Supporting Information.

I

5

88.2 Da, attributable to the cationic Ag complex

À

+

107

109

[

C H AgF NPSb (33)ÀSbF ] with the Ag and Ag iso-

33 37

6

topes. The measured mass distribution was in good agreement

+

with the simulated isotopic distribution for [C H AgNP] (for

33

38

chiometric amount of 1 and 2 and then stirred for 16 h at

details see Figure S70 (top) in the Supporting Information). An-

other smaller cluster of ions was observed with major positive

room temperature in CH Cl (Scheme 9), transparent solutions

2

I

were obtained. In the first reaction (Scheme 9, top), a layer of

n-hexane was carefully added, followed by standing at À308C.

Under these conditions colourless crystals of complex 33 were

obtained in a yield of 93% (see the Experimental Section). The

second experiment was aimed at isolating the silver(I) complex

of Danishefsky’s diene (31; Scheme 9, middle) from the result-

ing yellow-red solution. However, the attempt was unsuccess-

peaks at 847.2 and 849.1 Da, attributable to the cationic Ag

À

+

complex [C H Ag F N P Sb (33) À(SbF ) À(imine) +Cl ]

66

74

2

12

109

2

6 2

2

107

with the Ag and Ag isotopes (for details see Figure S70

(bottom) in the Supporting Information), which suggests that

under these conditions some CH Cl reacted, releasing chloride

2

that reacted to form the neutral dichloride-bridged disilver(I)

complex [{L1–Ag(m-Cl)} ] (18) characterised previously. When

2

I

ful and a mixture of cationic aquo–Ag–L1 complex 4 (see the

one chlorine atom is lost from 18, the major positive MS peaks

appear at 847.2 and 849.1 Da (see detailed fragmentation of

this cluster of peaks in Figure S70 (bottom) in the Supporting

Information). Finally, the catalytic activity of the cationic silver(I)

complex 33 in the two-component aza-Diels–Alder reaction

(Table 3, entry 7) was as high as that of the initial pre-catalyst

Experimental Section, Table S1 and Figures S3–S6 in the Sup-

porting Information) and a yellow-red oil that could be separ-

ated by crystallisation was obtained.

The [(L1)Ag(imine)][SbF ] complex 33, which corresponds to

6

I

the complex of the Ag pre-catalyst with imine 30, was isolated

I

and characterised by elemental analysis and spectroscopic

data (see the Experimental Section and Figures S65–S70 and

Table S9 in the Supporting Information for X-ray data and crys-

tal-packing details). The combustion elemental analysis data of

complex 33 is in accordance with the percentages expected

for its formula. Single crystals of complex 33 of sufficient qual-

ity for X-ray crystallography studies were obtained. The bond

lengths of Ag1ÀN1, Ag1ÀP1 and Ag1ÀC7_i_p_s_oand the distance

between the silver atom and the closest carbon atom (C8) of

the distal phenyl ring in complex 33 are 2.180(4), 2.3669(11),

Ag complexes 29, 27 or 10.

A mixture of aniline (8) and benzaldehyde (35) was used as

substrate instead of imine 30 to react with Danishefsky’s diene

(31) in a tandem three-coupling version of the aza-Diels–Alder

reaction catalysed by 6% of the neutral L2–Ag–(p-tolylsulfon-

I

ate) (29) or cationic L1–aniline–Ag (10) silver(I) complexes in

dioxane/water (3:1) under the optimised conditions (Table 3,

middle scheme). Under these conditions, yields of 50 and 45%

of 1,2-diphenyl-2,3-dihydro-4-pyridone (32) were formed by

using 29 and 10 as catalysts (Table 3, entries 8 and 9, respect-

ively) as well as an unknown product with a molecular mass at

161 Da (for GC-MS data see Figure S71 in the Supporting Infor-

2

.929 and 3.006 , respectively. These values are close to those

determined for complexes [(L1)Ag(pyrrolidine)][SbF ] (11) and

6

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

349

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

mation). This unknown by-product was present even when all

the imine 30 formed in situ was fully converted (for the GC-MS

data of imine 30 see Figure S72 in the Supporting Informa-

tion). However, slow addition of Danishefsky’s diene (31) to

the mixture of aniline (8) and benzaldehyde (35) over 50 min

using cationic 10 or 33 or neutral 29 or 27 silver(I) complexes

as catalyst improved the yield for this three-component reac-

tion, giving values that are comparable to those of the two-

component catalytic reaction (see Table 3, entries 10–13).

An alternative catalytic reaction was carried out in which

1

5 mol% of silver(I) complex 7 was added to aniline (8) at

room temperature and, after 1 h, the mixture was analysed by

3

1

P NMR spectroscopy, which showed the presence of two

31

doublets of P with a chemical shift at d=47.35 ppm with

1

107

31

1

109

31

J( Ag- P)=639.17 Hz and J( Ag- P)=738.97 Hz character-

I

istic of (L1–aniline–Ag ) complex 10 instead of two doublets of

31

1

107

31

P

109

with d=45.59 ppm with J( Ag- P)=691.53 Hz and

1

31

I

J( Ag- P)=798.39 Hz characteristic of (L1–toluene–Ag ) com-

plex 7 (see Figures S73 and S9 in the Supporting Information).

Next, an excess of benzaldehyde (35, 1.5 equiv) with respect to

the starting aniline (8) was added at room temperature and,

31

after 1 h, the P NMR spectrum was recorded, which showed

that complex 10 (dd, d=47.35 ppm) was completely con-

I

sumed and complex (L1–imine–Ag) 33 was formed, based on

3

1

the presence of two doublets of P at d=46.03 ppm with

1

107

31

1

109

31

J( Ag- P)=634.08 Hz and J( Ag- P)=732.02 Hz (see Fig-

ure S74 in the Supporting Information). After observation of

the complete formation of 33, at room temperature an excess

of Danishefsky’s diene (31, 2 equiv) with respect to the starting

aniline (8) was added and, after 30 min, the P NMR spectrum

was recorded. It was observed that complex 33 (dd, d=

Scheme 10. Mechanistic proposal for the three-component aza-Diels–Alder

cycloaddition leading to 1,2-diphenyl-2,3-dihydro-4-pyridone (32) using cat-

ionic Ag complex 7 or 11 as pre-catalyst in dioxane/water (1:1). Silver(I)

complexes 10 and 33 have been isolated as intermediates, compounds 37

and 38 have been characterised and silver(I) complexes 36, 39 and 40 are

proposed as intermediates.

3

1

I

3

1

4

6.03 ppm) was totally consumed and two new doublets of P

1

107

31

1

109

at d=47.21 ppm with J( Ag- P)=671.09 Hz and J( Ag-

31

P)=773.94 Hz, attributable to the presumed (L1–dihydropyr-

idin-4-one–Ag) complex 40, appeared together with two

doublet signals of (L1–aniline–Ag ) complex 10 (see Figure S75

I

presumed p-complex intermediate 36 formed between silver

[

17]

in the Supporting Information). For detailed measurements of

these coupling constants, see Figures S76 and S77, respective-

ly, in the Supporting Information. Over time (2 h), the peaks

corresponding to 40 and 10 were consumed with the

concomitant formation of 1,2-diphenyl-2,3-dihydro-4-pyridone

and benzaldehyde (35). The condensation reaction between

benzaldehyde (35) and aniline (8) takes place leading to the

formation of isolated silver(I) imine complex 33 and the release

of one molecule of water. The partial hydrolysis of Danishef-

sky’s diene (31) leads to the formation of trimethylsilanol (37)

and 4-methoxybut-3-en-2-one (38), which were characterised

by GC-MS (see Figure S79 for 37 and Figure S80 for 38 in the

Supporting Information). The presumed formation of p-com-

plex intermediate 39 between silver(I) complex 33 and silyl

enol ether 31, followed by cyclisation of the six-membered

ring would lead to the formation of intermediate L1–dihydro-

(

32). At the end of the reaction, the only silver(I) complex pres-

ent in the mixture was the cationic aquo–silver(I) complex 4,

3

1

as confirmed by the appearance in the P NMR spectra of two

1

107

31

doublets at d=46.08 ppm with J( Ag- P)=714.28 Hz and

1

109

31

J( Ag- P)=824.42 Hz (see Figure S78 in the Supporting In-

formation), due to the conditions used for this experiment in

I

dioxane/water (3:1), namely excess of Danishefsky’s diene (31,

pyridin-4-one–Ag complex 40 (see Scheme 10 and Figure S75

2

equiv) and benzaldehyde (35, 1.5 equiv) with respect to anil-

ine (8).

In accord with these experimental results, we propose

in the Supporting Information for NMR spectra experiment).

The release of 1,2-diphenyl-2,3-dihydro-4-pyridone (32) from

the coordination sphere of the metal ion by replacement with

free aniline (8) or imine 30 (for GC-MS characterisation of 30,

see Figure S41 in the Supporting Information) will initiate

a new catalytic cycle.

a plausible reaction mechanism for the three-component

coupling reaction with [Ag(L1)(toluene)][SbF₆] (7) or

[

(

Ag(L1)(pyrrolidine)][SbF ] (11) as pre-catalyst in dioxane/water

3:1; Scheme 10). The main feature of this mechanism is the

6

I

coordination of Ag complex 7 or 11 with aniline (8) leading to

the isolated intermediate [Ag(L1)(aniline)][SbF ] (10) and the

6

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

350

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

solved in dried dichloromethane (2 mL). The solution was stirred at

room temperature under an atmosphere of argon for 8 h. Then the

solvent was removed under reduced pressure, followed by

crystallisation in a mixture of non-dried dichloromethane/n-hexane

Conclusion

This study has provided an easy access to a series of cationic

and neutral silver(I) complexes bearing Buchwald-type phos-

phane ligands and an additional ligand exhibiting varied elec-

tronic and steric properties. These cationic and neutral silver(I)

complexes exhibit higher activity than analogous copper(I) and

gold(I) complexes as catalysts for the single and double Man-

(

1:3) at À308C for 24 h. [Ag(L1)(H O)][SbF ] (4) was collected

2

6

0.139 g, 84% yield) as colourless crystals suitable for X-ray crystal-

lography (see Table S1 and Figure S6 in the Supporting Informa-

1

tion). H NMR (300 MHz, CD Cl ): d=7.88 (t, 1H; ArH), 7.66–7.44 (m,

2

5H; ArH), 7.35–7.16 (m, 3H; ArH), 2.80 (br, 2H; H O), 1.29 (s, 9H;

2

3

nich A and two- and three-component aza-Diels–Alder coup-

tert-butyl CH

3

), 1.24 ppm (s, 9H; tert-butyl CH

) (see Figure S3 for

3

31

1

107

31

details);

14.28 Hz, J( Ag- P)=824.42 Hz) (see Figure S4 for details); MS

P NMR (CD Cl ): d=46.08 ppm (dd, J( Ag- P)=

ling reactions. A strong influence of the solvent on the activity

of the catalysts, attributable to limited solubility of the com-

plexes or their decomposition, has been observed. The isola-

tion of some silver(I) complexes that are likely to be reaction

intermediates has shed light on the reaction mechanisms and

provided a greater understanding the different stabilities of

2

1

109

31

7

À

+

(

[

(

ESI, +MS): m/z: 463.1 [C H AgF OPSbÀSbF +K] , 405.1

20

28

6

À + 107 109

C H AgF OPSbÀSbF ÀH O] with the Ag and Ag isotopes

20

28

6

2

À

see Figure S5 for details); MS (ESI, ÀMS): m/z: 234.9, 236.7 [SbF ]

6

121

123

with the Sb and Sb isotopes; elemental analysis calcd (%) for

C H AgF OPSb: C 36.45, H 4.28; found: C 36.56, H 4.35.

20

28

6

I

the Ag complexes compared with analogous Cu and Au com-

3

plexes. The mechanism of the A coupling reactions with the

Isolation of [Ag(L1)(PhCH )][SbF ] (7)

3

6

I

Ag complexes as catalysts presents similarities to the mechan-

+

À

I

A mixture of [Ag] [SbF ] (1; 0.086 g, 0.25 mmol) and 2-(di-tert-

ism with Cu complexes and differences to the mechanism

6

I

butylphosphino)biphenyl (L1=2; 0.075 g, 0.25 mmol) was dis-

solved in dried dichloromethane (2 mL). The solution was stirred at

room temperature under an atmosphere of argon for 8 h. Then the

solvent was removed under reduced pressure, followed by

crystallisation in a mixture of dried dichloromethane/toluene (1:2)

at room temperature for 48 h. [Ag(L1)(H O)][SbF ] (7) was collected

with Au complexes

Experimental Section

General

2

6

(

0.162 g, 88% yield) as colourless crystals suitable for X-ray crystal-

1

Details of the preparation, isolation and full characterisation of sil-

ver(I) complexes 4 and 7, the new silver(I) complexes 10, 11, 17,

lography (see Table S2 and Figure S7 for details); H NMR

[3]

(

300 MHz, CD Cl ): d=7.85 (t, 1H; ArH), 7.70–7.45 (m, 5H; ArH),

2 2

1

8, 27, 29 and 33, the phosphonium salt 5, mono-propargylamines

6, 21 and 24 and bis-propargylamines 22 and 25 c and 32, in-

7

.35–7.00 (m, 7H; ArH), 2.30 (s, 3H; CH of toluene), 1.23 (s, 9H;

3

[

12]

tert-butyl CH ), 1.17 ppm (s, 9H; tert-butyl CH ) (see Figure S8 for

3

31

1

107

31

cluding the NMR, ESI-MS and GC-MS spectra, combustion analysis

and single-crystal X-ray crystallography data are provided in the

Supporting

details);

91.531 Hz, J( Ag- P)=798.389 Hz). (see Figure S9 for details);

elemental analysis calcd (%) for C H AgF PSb: C 44.17, H 4.81;

P NMR (CD Cl ): d=45.59 ppm (dd, J( Ag- P)=

2 2

1

109

31

6

Information.

1414013 (11),

CCDC 1414010 (4),

1414014 (17),

1414011 (7),

1414015 (18),

27

35

6

1

414012 (10),

found: C 44.33, H 4.90.

414016 (27), 1414017 (29) and 1414018 (33) contain the supple-

mentary crystallographic data for this paper. These data are provid-

ed free of charge by The Cambridge Crystallographic Data Centre.

Isolation of [Ag(L1)(H NPh)][SbF ] (10)

2

6

+

À

A mixture of [Ag] [SbF ] (1; 0.086 g, 0.25 mmol) and 2-(di-tert-

6

All reactions were carried out under argon atmosphere in dried

1

solvent using a commercial solvent purification system. H NMR

butylphosphino)biphenyl (L1=2; 0.075 g, 0.25 mmol) and aniline

(8) (0.025 g, 0.27 mmol) was dissolved in dichloromethane (2 mL).

The solution was stirred at room temperature for 18 h under an at-

mosphere of argon. Then the resulting mixture was diluted with

spectra were recorded on a Bruker AV 300 MHz spectrometer.

1

Chemical shifts of the H signals are reported in ppm using the

solvent peak as the internal standard (CHDCl : d=5.27 ppm). Data

2

are reported as follows: chemical shift, multiplicity (s=singlet, br=

broad, d=doublet, dd=doublet of doublets, t=triplet, tt=triplet

of triplets, sept=septuplet, m=multiplet), coupling constants (Hz),

additional CH Cl (2 mL), filtered and the supernatant was covered

2 2

carefully with a layer of n-pentane (2 mL). Colourless crystals of 10

suitable for X-ray crystallography (see Table S3 and Figure S10 for

details) were obtained by standing for 48 h at À308C, and collect-

ed by filtration, washed with cold n-pentane and dried under

1

3

integral, and assignment. Chemical shifts of the C signal are also

reported in ppm using the solvent peak as the internal standard

3

1

(

CD Cl : d=53.84 ppm).

P NMR spectra were recorded on

vacuum to yield the corresponding complex [Ag(L1)(H

(10) (0.164 g, 89% yield). H NMR (300 MHz, CD Cl

2 2

2

NPh)][SbF

): d=7.85 (t, 1H;

ArH), 7.62–7.40 (m, 5H; ArH), 7.38–7.14 (m, 5H; ArH), 7.02 (t, 1H;

ArH), 6.93–6.81 (m, 2H; ArH), 4.20 (s, 2H; H NPh), 1.20 (s, 9H; tert-

) (see Figure S11 for de-

6

]

2

1

a Bruker 300 MHz spectrometer. Chemical shifts are reported in

ppm and coupling constants in Hz. Gas chromatography (GC) was

performed on a Varian 3900 apparatus equipped with a TRB-5MS

column (5% phenyl, 95% polymethylsiloxane, 30 m0.25 mm

2

butyl CH

tails);

), 1.15 ppm (s, 9H; tert-butyl CH

3

1

3

1

107

31

0

.25 mm, Teknokroma). GC-MS analyses were performed on an Agi-

P NMR (CD

2

Cl ): d=47.35 ppm (dd,

2

J( Ag- P)=

1

109

31

lent spectrometer (5973N-6890N) equipped with the same column

as the GC and operated under the same conditions. ESI-MS was

performed on an Agilent Esquire 6000 instrument. Elemental ana-

lyses were performed on an EuroEA Elemental Analyser Eurovector.

639.171 Hz, J( Ag- P)=738.974 Hz) (see Figure S12 for details);

13

C NMR (75 MHz, CD Cl ): d=149.45 (d), 149.20 (d), 143.03 (d),

2 2

142.91 (d), 141.81, 134.10 (d), 132.49 (d), 131.38 (d), 130.12, 129.99,

129.46, 128.15 (dd), 127.90, 126.73 (d), 126.33 (d), 124.56, 119.89,

3

5.65 (d), 35.46 (d), 30.88 (d), 30.75 ppm (d) (see Figure S13 for de-

13

tails); C DEPT-135 NMR (75 MHz, CD Cl ): negative signals: none;

2

Isolation of [Ag(L1)(H O)][SbF ] (4)

2

6

positive signals: d=134.10 (d), 132.49 (d), 131.38 (d), 130.12,

129.99, 129.47, 128.15 (d), 127.90, 124.56, 119.89, 30.88 (d),

30.75 ppm (d) (see Figure S14); MS (ESI, +MS): m/z: 405.1, 407.1

+

À

A mixture of [Ag] [SbF ] (1; 0.086 g, 0.25 mmol) and 2-(di-tert-

6

butylphosphino)biphenyl (L1=2; 0.075 g, 0.25 mmol) was dis-

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

351

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

À

+

[

C H AgF NPSbÀSbF Àaniline] ,

437.2

Isolation of [Ag(L1)(propargylamine)][SbF ] (17)

6

2

6

33

6

À

+

107

109

C H AgF NPSbÀSbF Àaniline+MeOH] for the Ag and Ag

2

6

33

6

+

À

A mixture of [Ag] [SbF ] (1; 0.086 g, 0.25 mmol), 2-(di-tert-butyl-

6

isotopes (see Figure S15 for details); MS (ESI, ÀMS): m/z: 234.9,

36.7 [SbF ] for the Sb and Sb isotopes (see Figure S15 for

À

121

123

phosphino)biphenyl (L1=2; 0.075 g, 0.25 mmol), pyrrolidine (9;

2

6

0

.019 g, 0.27 mmol), formaldehyde (15; aq. 37%) [0.024 g,

.30 mmol] and phenylacetylene (3) (0.025 g, 0.25 mmol) was sus-

details); elemental analysis calcd (%) for C H AgF NPSb: C 42.54,

H 4.53, N 1.91; found: C 42.76, H 4.80, N 2.1.

2

6

33

6

pended in toluene (2 mL). The mixture was stirred at room

temperature for 20 h under an atmosphere of argon. Then the pre-

cipitate formed in the resulting mixture was dissolved by adding

dichloroethane/dichloromethane (1:2, 3 mL), filtered and then al-

lowed to evaporate slowly at room temperature. Colourless crystals

of 17 suitable for X-ray crystallography (see Table S5 and Fig-

ure S28 for details) were collected by filtration, washed with cold

n-pentane and dried under vacuum to yield the corresponding

Isolation of [Ag(L1)(pyrrolidine)][SbF ] (11)

6

+

À

A mixture of [Ag] [SbF ] (1; 0.086 g, 0.25 mmol) and 2-(di-tert-

6

butylphosphino)biphenyl (L1=2; 0.075 g, 0.25 mmol) and pyrrol-

idine (9) (0.019 g, 0.27 mmol) was suspended in toluene (2 mL).

The solution was stirred at room temperature for 18 h under an at-

mosphere of argon. Then the precipitate formed in the resulting

mixture was dissolved by adding 1,2-dichloroethane (ClCH CH Cl;

complex [Ag(L1)(propargylamine)][SbF ] (17) (0.186 g, 90% yield).

6

1

H NMR (300 MHz, CD Cl ): d=7.84 (t, 1H; ArH), 7.69–7.45 (m, 5H;

2

ArH), 7.42–7.17 (m, 8H; ArH), 3.59 (m, 2H; CH ), 2.88 (m, 4H; (CH )-

2 2

2

mL), filtered and then a layer of n-pentane was carefully added.

pyrrolidine), 1.86 (m, 4H; (CH )-pyrrolidine), 1.20 (s, 9H; tert-butyl

2

Colourless crystals of 11 suitable for X-ray crystallography (see

Table S4 and Figure S16 for details) were obtained by slow evapor-

ation at room temperature, and were collected by filtration,

washed with cold n-pentane and dried under vacuum to yield the

CH ), 1.15 ppm (s, 9H; tert-butyl CH ) (see Figure S24 for details);

3

31

1

107

31

P NMR (CD Cl ): d=45.53 ppm (dd, J( Ag- P)=628.632 Hz,

2

1

109

31

13

J( Ag- P)=725.041 Hz) (see Figure S25 for details); C NMR

75 MHz, CD Cl ): d=149.35 (d), 142.79 (d), 134.15 (d), 132.66 (d),

(

2

corresponding complex ^[^A^g⁽^L¹⁾⁽^p^y^r^r^o^lⁱ^dⁱⁿ^e⁾^]^[^S^b^F6^](11; 0.163 g,

132.43, 131.51, 129.92, 129.70, 129.41, 129.04, 128.56, 128.45,

1

9

7

1% yield). H NMR (300 MHz, CD Cl ): d=7.87 (t, 1H; ArH), 7.64–

2

128.16, 125.64, 121.70, 88.17, 84.07, 58.80, 48.73, 35.72 (d), 35.74

.46 (m, 5H; ArH), 7.30–7.18 (m, 3H; ArH), 2.82 (m, 4H; (CH₂)-

13

(d), 31.04 (d), 30.90 (d), 23.36 ppm (see Figure S26 for details);

C

^p^y^r^r^o^lⁱ^dⁱⁿ^e⁾^,²^.³⁴⁽^b^r^,¹^H^;⁽^N^H⁾^-^p^y^r^r^o^lⁱ^dⁱⁿ^e⁾^,¹^.⁷¹⁽^m^,⁴^H^;⁽^C^H2⁾^-

DEPT-135 NMR (75 MHz, CD Cl ): negative signals: d=58.87 and

2

pyrrolidine), 1.29 (s, 9H; tert-butyl CH ), 1.24 ppm (s, 9H; tert-butyl

3

23.36 (CH , pyrrolidine), 48.79 (CH ); positive signals: d=134.16 (d),

2 2

3

1

CH ) (see Figure S17 for details); P NMR (CD Cl ): d=45.50 ppm

3

2

132.66 (d), 132.44, 131.51, 129.95, 129.71, 129.40, 129.06, 128.47,

28.19, 128.17, 30.98 ppm (d) (see Figure S27); elemental analysis

calcd (%) for C H AgF NPSb: C 47.91, H 5.12, N 1.69; found: C

1

107

31

1

109

31

(

dd, J( Ag- P)=608.486 Hz, J( Ag- P)=701.649 Hz) (see Fig-

ure S18 for details); C NMR (75 MHz, CD Cl ): d=149.55 (d),

1

3

2

33

42

6

1

3

49.30 (d), 142.85 (d), 142.73 (d), 134.09 (d), 132.49 (d), 131.44 (d),

29.77, 129.54, 128.20, 128.13 (d), 126.49 (d), 126.08 (d), 50.95,

5.67 (d), 35.51 (d), 31.11 (d), 30.98 (d), 24.45 ppm (see Figure S19

4

7.98, H 5.26, N 1.75.

1

3

Isolation of neutral [{L1-Ag(m-Cl)} ] (18)

for details); C DEPT-135 NMR (75 MHz, CD Cl ): negative signals:

2

d=50.95 and 24.45 (CH , pyrrolidine); positive signals: d=134.10

2

The ligand 2-(di-tert-butylphosphino)biphenyl (L1=2; 0.074 g,

(

3

d), 132.49 (d), 131.44 (d), 129.77, 129.54, 128.20, 128.14 (d), 50.95,

1.11 (d), 30.98 (d), 24.45 ppm (see Figure S20); elemental analysis

calcd (%) for C H AgF NPSb: C 40.42, H 5.29, N 1.96; found: C

0

.25 mmol) was added to a suspension of AgCl (0.043 g, 0.3 mmol)

in dichloromethane (4 mL). The suspension was stirred for 48 h at

08C, filtered to remove unreacted AgCl and after evaporation of

2

4

36

6

5

4

0.60, H 5.51, N 2.03.

the solvent under reduced pressure at 458C, the solid residue was

washed with cold pentane (32 mL) and dried to afford the prod-

uct as a colourless solid (0.195 g, 88%). Colourless crystals of 18

suitable for X-ray crystallography (see Table S6 and Figure S37 for

details) were obtained by slow evaporation of the resulting mix-

3

General procedure for the single Mannich A coupling

reaction

ture in dichloromethane after standing for 4 h at room tempera-

1

Preparation of propargylamine 16: A mixture of phenylacetylene

ture. H NMR (300 MHz, CD

2

Cl ): d=7.87 (t, 2H; ArH), 7.55–7.40 (m,

2

(

0

(

0.25 mmol), pyrrolidine (0.35 mmol), formaldehyde (aq. 37%,

.7 mmol) and silver(I) complexes 7, 10, 11, 17, 27 or 29

0.015 mmol) as catalyst (Ag/alkyne ratio 6 mol%) was suspended

in toluene or CH Cl (1 mL). Then the flask was evacuated under

10H; ArH), 7.30–7.21 (m, 2H; ArH), 7.19–7.10 (m, 4H; ArH), 1.30 (s,

18H; tert-butyl CH ), 1.25 ppm (s, 18H; tert-butyl CH ) (see Fig-

3

31

1

107

ure S33 for details); P NMR (CD

Cl

): d=43.68 ppm (dd, J( Ag-

2

31

1

109

31

P)=605.605 Hz, J( Ag- P)=698.304 Hz) (see Figure S34 for de-

2

13

vacuum and refilled with argon. The evacuation/refilling cycle was

repeated three times (pressure 2 bar). The mixture was stirred at

room temperature for 3–6 min to obtain the maximum yield of 16

tails); C NMR (75 MHz, CD Cl ): d=150.56 (d), 150.30 (d), 142.08

2 2

(d), 141.97 (d), 134.15 (d), 132.51 (d), 130.87 (d), 129.81, 128.89,

128.83, 127.66 (d), 127.51 dd), 127.29 (d), 35.66 (d), 35.51 (d), 31.18

13

(

see Table 1). Then cold n-hexane was added to the cold reaction

(d), 31.05 ppm (d) (see Figure S35 for details); C DEPT-135 NMR

(75 MHz, CD Cl ): negative signals: none; positive signals: d=

mixture, which was filtered and the solvents were removed under

reduced pressure at 508C to provide propargylamine 16 as a light-

yellow oil in a yield of 98%. The structure and purity of propargyl-

amine 16 was confirmed by H and C NMR spectroscopy as well

as by GC-MS analysis:

2

134.25 (d), 132.52 (d), 130.87 (d), 129.81, 128.89, 128.83, 127.51

(dd), 31.19 (d), 31.05 ppm (d) (see Figure S36); elemental analysis

1

13

calcd (%) for C40

6.25.

H54Ag

Cl

P

: C 54.38, H 6.16; found: C 54.42, H

2

1

Compound 16: H NMR (300 MHz, CD Cl ): d=7.40–7.20 (m, 5H;

2

C H ), 3.54 (s, 2H; CH ), 2.59 (m, 4H; CH -pyrrolidine), 1.74 ppm (m,

3

6

5

2

1

General procedure for the Mannich double A coupling

reaction

3

4

H; CH -pyrrolidine) (see Figure S21); C NMR (75 MHz, CD Cl ):

2 2 2

d=131.97, 128.67, 128.33, 123.79, 86.32, 84.25, 52.94, 44.04,

2

4.24 ppm (see Figure S22); GC-MS: m/z: 185.2 [C H N] (see Fig-

Preparation of bis-propargylamines 22 and 25: A mixture of 1,4-

diethynylbenzene (20) or 1,6-heptadiyne (23; 0.25 mmol), pyrrol-

1

3

15

ure S23 for details).

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

352

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

I

idine (0.70 mmol), formaldehyde (aq. 37%, 1.4 mmol) and Ag com-

129.14, 128.90, 128.57, 127.70 (dd), 126.41, 30.97 (d), 30.84 (d),

plex 11 (0.015 mmol) as catalyst (Ag/alkyne ratio 6 mol%) was sus-

21.45 ppm (see Figure S52); MS (ESI, +MS): m/z: 405.1, 407.1

À

+

107

109

pended in CH Cl (1 mL) and then the flask was evacuated under

[C H AgO PSÀ(OTf) ] for the Ag and Ag isotopes (see Fig-

2

27 34

3

À

vacuum and refilled with argon. The evacuation/refilling cycle was

repeated three times (pressure 2 bar). The mixture was stirred at

room temperature for the required time to obtain the maximum

yield of 22 or 25 (see Table 2). Then cold n-hexane was added to

the cold reaction mixture, which was filtered and the solvents re-

moved under reduced pressure at 508C to provide the bis-propar-

gylamines as a light-yellow solid (22) and light-yellow oil (25) in

yields in excess of 90%. The structures and purity of bis-propargyl-

ure S54 for details); MS (ESI, ÀMS): m/z: 170.08 [C H O S] (see Fig-

7

3

ure S54 for details); elemental analysis calcd (%) for

C H AgO PS·CH Cl : C 49.43, H 5.63, S 4.71; found: C 49.54, H

27

36

4

2

5.69, S 4.56.

Isolation of [Ag(L2)(p-tolylsulfonate)] (29)

I

A mixture of Ag p-tolylsulfonate (26) (0.056 g, 0.25 mmol) and 2-

1

13

amines 22 and 25 were confirmed by H, C and C DEPT-135

NMR spectroscopy as well as by GC-MS analysis. The correspond-

ing primary mono-adducts intermediates 21 and 24 were also de-

tected in the reaction mixture and were characterised by GC-MS of

the crude reactions:

(

di-tert-butylphosphanyl)(2’,4’,6’-triisopropyl)biphenyl

(L2=28;

0

.106 g, 0.25 mmol) was dissolved in dichloromethane (2 mL). The

mixture was stirred at room temperature for 24 h under an atmos-

phere of argon. Then the resulting mixture was diluted by adding

dichloromethane (2 mL), filtered and then the supernatant layered

carefully with n-hexane (2 mL). Colourless crystals of 29 suitable for

X-ray crystallography (see Table S8 and Figure S59 for details) were

obtained by standing for 48 h at À308C. The crystals were collect-

ed by filtration and dried under vacuum to yield the corresponding

1

Compound 22: H NMR (300 MHz, CD Cl ): d=7.30 (s, 4H; C H ),

2

6

4

3

.56 (s, 4H; CH ), 2.61 (m, 8H; CH -pyrrolidine), 1.75 ppm (m, 8H;

2 2

1

3

CH -pyrrolidine) (see Figure S38). C NMR (75 MHz, CD Cl ): d=

2

1

31.89, 123.33, 87.78, 84.14, 52.85, 43.99, 24.25 ppm (see Fig-

1

3

ure S39); C DEPT-135 NMR (75 MHz, CD Cl ): negative signal: d=

1

2

complex [Ag(L2)(p-tolylsulfonate)] (29; 0.161 g, 92% yield). H NMR

1

4

31 (CH, C H ); positive signals: d=52.85, 24.25 (CH , pyrrolidine),

6 4 2

(300 MHz, CD Cl ): d=7.86 (t, 1H; ArH), 7.64–7.55 (m, 2H; ArH),

2

3.99 ppm (CH , group) (see Figure S40); GC-MS: m/z: 292.2

2

7.54–7.42 (m, 2H; ArH), 7.35–7.25 (m, 1H; ArH), 7.16–7.06 (m, 4H;

ArH), 2.78 (sept, 1H; CH-isopropyl), 2.33 (sept, 2H; CH-isopropyl),

2.31 (s, 3H; CH -Ar), 1.28 (s, 9H; tert-butyl CH ), 1.24 (d, 6H; CH -

[

C H N ] (see Figure S41 for details); elemental analysis calcd (%)

20 24 2

for C H N ·H O: C 77.38, H 8.44, N 9.02; found: C 76.74, H 8.63, N

2

0

24

2

3

8

.67.

isopropyl), 1.23 (s, 9H; tert-butyl CH ), 1.19 (d, 6H; CH -isopropyl),

3

Single coupling intermediate 21: GC-MS: m/z: 209.2 [C H N] (see

Figure S42 for details).

0.87 ppm (d, 6H; CH -isopropyl) (see Figure S55 for details);

1

5

15

3

31

1

107

31

P NMR (CD Cl ): d=42.09 ppm (dd, J( Ag- P)=685.329 Hz,

2 2

1

109

31

13

1

4

J( Ag- P)=791.345 Hz) (see Figure S56 for details); C NMR

75 MHz, CD Cl ): d=150.84, 147.91 (d), 147.65 (d), 146.11, 142.54,

Compound 25: H NMR (300 MHz, CD Cl ): d=3.25 (t, J=2.20 Hz,

2

a

4

2

(

4

7

H; CH ), 2.48 (m, 8H; CH -pyrrolidine), 2.24 (tt, J=2.20, J=

2

b

1

40.55, 135.15 (d), 134.49 (d), 134.37 (d), 130.65 (d), 128.99, 127.43

.03 Hz, 4H; CH ), 1.69 (m, 4H; CH -pyrrolidine), 1.61 (quint, 2H;

2

13

c

(

2

dd), 126.44, 122.61, 31.38, 31.29 (d), 31.16 (d), 26.51, 23.85, 23.12,

1.42 ppm (see Figure S57 for details); C DEPT-135 NMR (75 MHz,

CH ), 0.83 ppm (see Figure S44); C NMR (75 MHz, CD Cl ): d=

2

13

8

3.40, 76.85, 52.81, 43.62, 28.67, 24.16, 18.09 ppm (see Figure S45);

1

3

CD Cl ): negative signals: none; positive signals: d=135.14 (d),

C DEPT-135 NMR (75 MHz, CD Cl ): positive signals: d=52.81,

2

1

3

33.91 (d), 130.64 (d), 128.99, 127.43 (dd), 126.43, 122.60, 34.37,

1.38, 31.28 (d), 31.16 (d), 26.51, 23.85, 23.12, 21.42 ppm (see

4

3.62, 28.66, 24.16, 18.09 ppm (CH , groups) (see Figure S46); GC-

2

MS: m/z: 258.2 [C H N ] (see Figure S47 for details).

1

7

26

2

Figure S58 for details); MS (ESI, +MS): m/z: 765.2

Single adduct intermediate 24: GC-MS: m/z: 175.2 [C H N] (see

Figure S48 for details).

1

2

17

+

À

+

[

C H AgO PS+2MeOH] , 531.4, 533.2 [C H AgO PSÀ(OTf) ] for

36

52

07

3

36 52

3

1

109

the Ag and Ag isotopes (see Figure S60 for details); elemental

analysis calcd (%) for C H AgO PS: C 61.45, H 7.45, S 4.56; found:

36

52

3

Isolation of [Ag(L1)(p-tolylsulfonate)(H O)] (27)

C 61.49, H 7.53, S 4.50.

2

I

A mixture of Ag p-tolylsulfonate (26) (0.056 g, 0.25 mmol) and 2-

(

di-tert-butylphosphino)biphenyl (L1=2; 0.075 g, 0.25 mmol) was

General procedure for the aza-Diels–Alder two- and three-

component coupling reactions

dissolved in dichloromethane (2 mL). The mixture was stirred at

room temperature for 24 h under an atmosphere of argon. Then

the resulting mixture was diluted by adding dichloromethane

Preparation of 1,2-diphenyl-2,3-dihydro-4-pyridone (32): Two-

component coupling: a mixture of imine 30 (0.25 mmol) and Dani-

(

2 mL), filtered and then the supernatant was layered carefully with

n-hexane (2 mL). Colourless crystals of 27 suitable for X-ray crystal-

lography (see Table S7 and Figure S53 for details) were obtained

by standing for 48 h at À308C. The crystals were collected by filtra-

shefsky’s diene (31; 0.3 mmol) was suspended in dioxane/H O (3/1)

(2 mL). Three-component coupling: Danishefsky’s diene (31) dis-

solved in dioxane (1 mL) was slowly added to a mixture of aniline

2

tion and dried under vacuum to yield the corresponding complex

(8; 0.25 mmol) and benzaldehyde (35; 0.25 mmol) suspended in di-

1

I

[

Ag(L1)(p-tolylsulfonate)(H O)] (27; 0.133 g, 92% yield). H NMR

oxane/H O (1:1; 1 mL) over 50 min. In both cases Ag complex 27,

2

(

7

9

300 MHz, CD Cl ): d=7.86 (t, 1H; ArH), 7.68–7.23 (m, 8H; ArH),

.22–7.12 (m, 4H; ArH), 2.33 (s, 3H; CH ), 1.59 (s, 2H; H O), 1.27 (s,

3 2

29 or 33 (0.015 mmol) was used as catalyst (Ag/alkyne ratio

6 mol%). Then the flask was evacuated under vacuum and refilled

with argon. The evacuation/refilling cycle was repeated three times

(pressure 2 bar). The mixtures were stirred at room temperature for

the required time to obtain the maximum yield of 32 (see Table 3).

Then the solvents were removed under reduced pressure at 608C.

The reaction mixture was diluted with water and the organic

material was extracted with CH Cl to provide 1,2-diphenyl-2,3-di-

2

H; tert-butyl CH ), 1.22 ppm (s, 9H; tert-butyl CH ) (see Figure S49

3

1

107

31

for details); P NMR (CD Cl ): d=43.93 ppm (dd, J( Ag- P)=

2

1

109

31

6

95.077 Hz, J( Ag- P)=802.180 Hz) (see Figure S50 for details);

C NMR (75 MHz, CD Cl ): d=150.15 (d), 149.90 (d), 142.38, 141.85

1

3

2

(

1

3

d), 141.73 (d), 140.82, 134.12 (d), 132.28 (d), 131.02 (d), 129.62,

29.14, 128.91, 128.57, 127.70 (dd), 127.27 (d), 126.89 (d), 126.42,

5.66 (d), 35.52 (d), 30.98 (d), 30.85 (d), 21.45 ppm (see Figure S51

2

hydro-4-pyridone (32) in a yield higher than 90%. The structure

1

3

1

13

for details); C DEPT-135 NMR (75 MHz, CD Cl ): negative signals:

and purity of 32 were confirmed by H, C and C DEPT-135 NMR

2

none; positive signals: d=134.12 (d), 132.28 (d), 131.02 (d), 129.61,

www.chemeurj.org

spectroscopy as well as by GC-MS analysis.

Chem. Eur. J. 2016, 22, 340 – 354

353

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

1

Compound 32: H NMR (300 MHz, CD Cl ): d=7.80 (d, J=1.2 Hz,

Gold, pp. 357–379, DOI:10.1002/9780470597521.ch12 ; b) H. V. R. Dias,

J. A. Flores, J. Wu, P. Kroll, J. Am. Chem. Soc. 2009 , 131 , 11249–11255;

c) S. Díez-Gonzµlez, S. P. Nolan, Acc. Chem. Res. 2008 , 41 , 349–358; d) M.

Rudolph, A. S. K. Hashmi, Chem. Soc. Rev. 2012 , 41 , 2448–2462; e) G.

Abbiati, E. Rossi, Beilstein J. Org. Chem. 2014 , 10 , 481–513 ; f) O. Prakash,

H. Joshi, U. Kumar, A. K. Sharma, A. K. Singh, Dalton Trans. 2015 , 44 ,

2

1

2

H), 7.38–7.10 (m, 10H), 7.08–6.98 (m, 3H), 5.26 (dd, J=0.9, 7.8 Hz,

H), 3.26 (dd, J=7.2, 16.5 Hz, 1H), 2.70 ppm (ddd, 1H) (see Fig-

ure S61 for details); C NMR (75 MHz, CD Cl ): d=190.83, 150.46,

44.80, 138.16, 129.95, 129.38, 128.34, 126.58, 125.32, 119.41,

01.98, 62.07, 43.52 ppm (see Figure S62); C DEPT-135 NMR

1

3

2

1

3

1962–1968; g) M. Trose, M. Dell’Acqua, T. Pedrazzini, V. Pirovano, E.

(

75 MHz, CD Cl ): negative signals: d=43.48 ppm; positive signals:

2 2

Gallo, E. Rossi, A. Caselli, G. Abbiati, J. Org. Chem. 2014 , 79 , 7311–7320 ;

h) N. Salam, A. Sinha, A. S. Roy, P. Mondal, N. R. Jana, S. M. Islam, RSC

Adv. 2014 , 4 , 10001–10012; i) C. Ameta, K. L. Ameta, Heterogeneous Cat-

alysis. A Versatile Tool for the Synthesis of Bioactive Heterocycles , CRC

Press, Boca Raton, 2015, pp. 303–320, DOI: 10.1201/b17418–12; j) H.

Mandai, K. Mandai, M. L. Snapper, A. H. Hoveyda, J. Am. Chem. Soc.

d=150.48, 129.95, 129.38, 128.37, 126.58, 125.31, 119.40, 102.00,

2.07 ppm (see Figure S63 for details); GC-MS: m/z: 249.12

C H NO] (see Figure S64 for details);

6

[

1

7

15

Isolation of [Ag(L1)(imine)][SbF ] (33)

6

2008 , 130 , 17961–17969; k) M. Kawasaki, H. Yamamoto, J. Am. Chem.

+

À

Soc. 2006 , 128 , 16482–16483.

2] a) L. R. Moore, S. M. Cooks, M. S. Anderson, H.-J. Schanz, S. T. Griffin,

R. D. Rogers, M. C. Kirk, K. H. Shaughnessy, Organometallics 2006 , 25 ,

A mixture of [Ag] [SbF ] (1; 0.086 g, 0.25 mmol), 2-(di-tert-butyl-

6

[

phosphino)biphenyl (L1=2; 0.075 g, 0.25 mmol) and imine 30

0.045 g, 0.25 mmol) was suspended in dried dichloromethane

2 mL). The mixture was stirred at room temperature for 16 h

under an atmosphere of argon. Then the transparent resulting so-

lution was carefully layered with n-pentane, followed by standing

for 48 h at À308C to afford colourless crystals of complex 33 suit-

able for X-ray crystallography (see Table S9 and Figure S69 for de-

tails). These crystals were collected by filtration, washed with cold

n-pentane and dried under vacuum to yield the corresponding

complex [Ag(L1)(imine)][SbF ] (33; 0.191 g, 93% yield). H NMR

300 MHz, CD Cl ): d=8.68 (s, 1H; (N=CH)-imine), 7.96–7.87 (m,

2 2

H; ArH), 7.81–7.64 (m, 3H; ArH), 7.61–7.37 (m, 7H; ArH), 7.28–7.05

(

5151–5158; b) M. L. Gallego, P. Ovejero, M. Cano, J. V. Heras, J. A.

Campo, E. Pinilla, M. R. Torres, Eur. J. Inorg. Chem. 2004 , 3089–3098 ;

c) S. S. Y. Chui, M. F. Y. Ng, C.-M. Che, Chem. Eur. J. 2005 , 11 , 1739–1749.

[3] P. PØrez-Galµn, N. Delpont, E. Herrero-Gomez, F. Maseras, A. M. Echavar-

ren, Chem. Eur. J. 2010 , 16 , 5324–5332, S5324/5321-S5324/5156.

[4] C. Wei, Z. Li, C.-J. Li, Org. Lett. 2003 , 5 , 4473–4475.

5] Z. Liu, P. Liao, X. Bi, Org. Lett. 2014 , 16 , 3668–3671.

6] C. Loncaric, K. Manabe, S. Kobayashi, Adv. Synth. Catal. 2003 , 345 , 475–

477.

7] C.-G. Yang, N. W. Reich, Z. Shi, C. He, Org. Lett. 2005 , 7 , 4553–4556.

8] a) M. Raducan, C. Rodriguez-Escrich, X. C. Cambeiro, E. C. Escudero-

Adan, M. A. Pericas, A. M. Echavarren, Chem. Commun. 2011 , 47 , 4893–

4895; b) A. Das, C. Dash, M. Yousufuddin, M. A. Celik, G. Frenking,

H. V. R. Dias, Angew. Chem. Int. Ed. 2012 , 51 , 3940–3943; Angew. Chem.

2012 , 124 , 4006–4009, S3940/3941-S3940/3925 ; c) P. Rçmbke, A. Schier,

H. Schmidbaur, J. Chem. Soc. Dalton Trans. 2001 , 2482–2486.

9] H. Yoshida, I. Kageyuki, K. Takaki, Org. Lett. 2014 , 16 , 3512–3515.

10] D. S. Surry, S. L. Buchwald, Angew. Chem. Int. Ed. 2008 , 47 , 6338–6361;

Angew. Chem. 2008 , 120 , 6438–6461.

11] a) L. W. Bieber, M. F. Da Silva, Tetrahedron Lett. 2004 , 45 , 8281–8283 ;

b) B. R. Buckley, A. N. Khan, H. Heaney, Chem. Eur. J. 2012 , 18 , 3855–

[

1

6

[

(

1

(

^m^,⁶^H^;^A^r^H⁾^,⁷^.⁰³^–⁶^.⁹²⁽^m^,²^H^;^A^r^H⁾^,¹^.²⁹⁽^s^,⁹^H^;^t^e^r^t^-^b^u^t^y^l^C^H3⁾^,

1

.24 ppm (s, 9H; tert-butyl CH ) (see Figure S65 for details);

3

P NMR (CD Cl ): d=46.03 ppm (dd, J( Ag- P)=634.082 Hz,

3

1

107

31

2

1

109

31

13

J( Ag- P)=732.025 Hz) (see Figure S66 for details); C NMR

75 MHz, CD Cl ): d=168.84, 149.82, 149.34 (d), 149.08, 142.73 (d),

42.62, 142.60, 134.73, 134.53, 134.03 (d), 132.73 (d), 131.68 (d),

30.51, 130.31, 129.82, 129.22, 128.87, 128.79, 128.42, 128.32,

26.07 (d), 125.65 (d), 122.58, 36.01 (d), 35.85 (d), 31.06 (d),

0.93 ppm (d) (see Figure S67 for details); C DEPT-135 NMR

75 MHz, CD Cl ): negative signals: none; positive signals: d=

68.82, 134.72, 134.00 (d), 132.70 (d), 131.66 (d), 130.48, 130.28,

29.79, 129.19, 128.84, 128.76, 128.30, 122.56, 31.02 (d), 30.89 ppm

d) (see Figure S68 for details); elemental analysis calcd (%) for

[

(

2

[

1

3

1

3

3858, S3855/3851-S3855/3825; c) Y. Zhang, D. Yu, in Method for Prepara-

(

I

2

tion of Propargylamines via Cu-Catalyzed Three Component Coupling Re-

action, Patent Number: WO2012005692A1; d) V. A. Peshkov, O. P. Pere-

shivko, E. V. Van der Eycken, Chem. Soc. Rev. 2012 , 41 , 3790–3807; e) W.-

J. Yoo, L. Zhao, C.-J. Li, Aldrichimica Acta 2011, 44, 43–51; f) C. Wei, Z. Li,

C.-J. Li, Synlett 2004, 1472–1483.

1

(

C H AgF NPSb: C 48.20, H 4.54, N 1.70; found: C 48.28, H 4.61, N

33

37

6

[

12] P. J. Alaimo, R. O’Brien III, A. W. Johnson, S. R. Slauson, J. M. O’Brien, E. L.

Tyson, A.-L. Marshall, C. E. Ottinger, J. G. Chacon, L. Wallace, C. Y. Pauli-

no, S. Connell, Org. Lett. 2008 , 10 , 5111–5114.

1

.75.

[

13] A. Grirrane, E. Alva rez, H. Garcia, A. Corma, Angew. Chem. Int. Ed. 2014 ,

Acknowledgements

5

3 , 7253–7258; Angew. Chem. 2014 , 126 , 7381–7386.

14] A. Grirrane, E. Alvarez, H. Garcia, A. Corma, Chem. Eur. J. 2014 , 20 ,

Financial support by the Spanish ministry of Economy and

Competitiveness (Severo Ochoa and CTQ2012-32315) and the

Generalidad Valenciana (Prometeo 2012-014) is gratefully ac-

knowledged.

14317–14328.

[15] a) A. Grirrane, H. Garcia, A. Corma, E. Alvarez, ACS Catal. 2011 , 1 , 1647–

1653 ; b) A. Grirrane, H. Garcia, A. Corma, E. Alvarez, Chem. Eur. J. 2013 ,

1

9 , 12239–12244.

[

16] A. Homs, I. Escofet, A. M. Echavarren, Org. Lett. 2013 , 15 , 5782–5785.

17] Y.-Y. Jiang, H.-Z. Yu, Y. Fu, Organometallics 2014 , 33 , 6577–6584.

Keywords: cross-coupling

·

homogeneous catalysis

·

P

ligands · reaction mechanisms · silver

[

1] a) A. S. K. Hashmi, in Silver in Organic Chemistry (Ed.: M. Harmata), Wiley,

Hoboken, 2010, Chapter 12: A Critical Comparison: Copper, Silver, and

Received: August 26, 2015

Published online on November 24, 2015

Chem. Eur. J. 2016, 22, 340 – 354

www.chemeurj.org

354

ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Article Doi

Catalytic Activity of Cationic and Neutral Silver(I)-XPhos Complexes with Nitrogen Ligands or Tolylsulfonate for Mannich and Aza-Diels-Alder Coupling Reactions

DOI: 10.1002/chem.201503386

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.201503386

Full text of DOI:10.1002/chem.201503386

Products guided by the article

R&D Labs maybe for 1015-02-7

Relevant to this article

Hot Product