Phosphorus functionalized dendrimers and hyperbranched polymers: Is there a need for perfect dendrimers in catalysis?

Israel Journal of Chemistry Vol. 49

DOI: 10.1560/IJC.49.1.79

2009

pp. 79–98

Phosphorus Functionalized Dendrimers and Hyperbranched Polymers:

Is There a Need for Perfect Dendrimers in Catalysis?

Fabrizio ribaudo, Piet W.N.M. vaN LeeuWeN, aNd Joost N.H. reek*

Supramolecular and Homogeneous Catalysis, van’t Hoff Institute for Molecular Sciences, University of Amsterdam,

Nieuwe Achtergracht 166, Amsterdam, 1018 WV, The Netherlands

(Received 21 July 2008 and in revised form 4 November 2008)

Abstract. In this paper we describe the facile and straightforward covalent func-

tionalization of commercially available dendritic poly(propylenimine) and hyper-

branched poly(ethylenimine) with P-containing functional groups. The P-function-

alized macromolecules have been applied as multivalent ligands in the Pd-catalyzed

allylic substitution reactions (batch and continuous process) using either morpholine

or thiophenol as nucleophile. Palladium complexes of all described molecules are

active in allylic substitution reactions. The PEI functionalized polymers appear more

sensitive to small changes in the P/Pd ratio than the PPI analogues, but form cata-

lysts that are more active. When used in a continuous ﬂow process the macromol-

ecules are completely retained by the nanoﬁltration membrane, while the catalytic

activity decreases with time because of palladium depletion. This is more severe

for the allylic thiolation, probably because of the stronger afﬁnity of sulfur for pal-

ladium, facilitating palladium leaching.

IntroductIon

sults were obtained by using in-situ prepared palladium

In the last decades several synthetic procedures have complexes of a 4th generation dendrimer (calculated

been developed for the large-scale preparation of (al- molecular weight 20,564 Da for 100% palladium load-

most) perfect dendritic and hyperbranched polymers. ing of the 32 diphosphines). After 100 residence times,

Vögtle, and later Mülhaupt and Meijer, described the the conversion had decreased from 100% to approxi-

divergent synthesis of poly(propylenimine) dendrimers mately 75%. The formation of inactive PdCl₂was pro-

from the diaminobutane core.¹This type of dendritic posed to account for an additional drop in conversion,

scaffold has already been used for covalent functional- besides the small amount that leaches.

ization with appropriate functional groups that can act

Kaneda and coworkers used similar components

as ligands in transition metal catalysis. Reetz et al. func- to prepare phosphinated dendrimer-bound Pd(0) com-

tionalized commercially available DAB-dendrimers plexes that were applied in allylic amination reactions.⁴

with diphenylphosphine groups at the periphery via The catalysts showed high stereoselectivity due to the

double phosphination of the amines with diphenylphos- surface congestion of the dendrimers and could be ef-

phine and formaldehyde.²The corresponding Pd com- ﬁciently recycled without loss of activity under thermo-

plexes have been prepared (Fig. 1) and used as catalyst morphic conditions.

for the Heck coupling reaction between bromobenzene

and styrene.

The structural perfection that may be present in a

dendritic support might not be required for every ap-

Brinkmann et al. reported the use of these phosphine- plication in catalysis, and hyperbranched polymers pro-

containing dendritic catalysts for allylic substitution vide interesting and cheap alternatives as catalyst sup-

reactions in a Continuous Flow Membrane Reactor.³ports. These hyperbranched polymers are obtained from

Retention factors up to 0.999 were reported for the 3rd a simple one-pot synthesis, yielding globular polymeric

generation dendrimer (calculated molecular weight = structures with broader molecular weight distributions

10212 Da). Although the dendrimer remained in the

reactor, leaching of palladium was observed. Better re- E-mail: reek@science.uva.nl

*Author to whom correspondence should be addressed.

80

Fig. 1. Metal complexes of phosphane-functionalized poly(propylenimine) dendrimers.

compared to their dendritic analogues. Poly(ethylene- even at high conversions (96%). Haag and coworkers

imine) in particular is prepared by cationic ring-opening prepared hyperbranched polyethylenimine with carbo-

polymerization of aziridine (ethylenimine), and it can hydrate terminal groups (glycidol, gluconolactone, or

be prepared at large scale. This compound has already lactobionic acid) and used them as stabilizing support

been known for decades but recently new synthetic pro- materials for colloidal metal nanoparticles (i.e., Cu, Ag,

cedures have been developed that lead to polymers with Au, and Pt) in water.⁹The platinum nanoparticles stabi-

a high degree of branching (~65%) and (relatively) nar- lized with PEI derivatives proved to be as effective and

row polydispersity (ca. 1.3 for the commercial PEI-5, robust for the selective hydrogenation of isophorone in

polyethylenimine with MW ~5000).⁵From a chemical water as the dendrimer derivatives.

point of view these polymers are closely related to the

It is clear from the contributions of many groups that

PPI-dendrimers as their backbones are only made of C dendrimers and hyperbranched polymers are suitable

and N. Importantly, the nature of the N atoms is differ- supports for transition metal catalysts. The two main as-

ent; only tertiary and primary nitrogen atoms are found pects related to the use of dendritic supports in catalysis

for the PPI-dendrimer, whereas the hyperbranched PEI are the so-called “dendritic effect” and recycling. The

also contains secondary amine groups. Another differ- special properties that may result from catalyst incor-

ence with respect to PPI dendrimers is the number of poration onto these macromolecules can be described

C atoms in the spacer of the repetitive unit, being 3 for as a “dendrimer effect”. This term has been gener-

PPI dendrimers and 2 for hyperbranched PEI. It is un- ally invoked in the literature to explain phenomena that

known if and how the different pattern of branching and arise as the generation of the dendrimer increases. At

the smaller repetitive unit will affect their properties as this stage it is important to note that there is no general

supports for catalysts. This class of polymers has been trend observed in the many reports concerning cataly-

used for antimicrobial properties, coating preparation, sis with periphery-functionalized dendrimers. In some

and gene delivery,⁶but reports on applications as sup- cases the activity of the catalysts decreases with the

ports for transition metal catalysis are still limited.⁷dendrimer generation, while in other examples it does

Frey and coworkers used modiﬁed hyperbranched not. As already mentioned, the synthesis of (perfect)

poly(ethylenimine) as macroligand for the copper-me- dendrimers can be very time-consuming and costly.

diated atom transfer radical polymerization (ATRP) of While the highly symmetric dendrimers give beauti-

methyl methacrylate (MMA).⁸This system appeared ful pictures that have often served as covers for many

to be catalytically efﬁcient, providing polymers with journals, one might wonder if the efforts to prepare

narrow molecular weight distributions (M_w/M_n< 1.4) these macromolecules are justiﬁed by the added value

Israel Journal of Chemistry

49

2009

81

they have in catalysis. Cheaper materials like the hy-

The coupling of 1 to the dendrimer via urea moieties

perbranched polymers represent interesting alternatives was accomplished following an adapted literature pro-

to be used as supports for recyclable catalysts. The cedure using 1,1¢-carbonyldiimidazole (CDI).¹¹This

price per active site of this type of materials competes reaction was ﬁrst optimized using a diamine that re-

with other supports such as polystyrene and silica. In sembles the peripheral diamino end-groups present on

this paper we describe new strategies for the covalent the DAB dendrimers (Scheme 2) before application of

functionalization of dendritic poly(propylenimine) and the synthetic strategy to the larger systems.

hyperbranched poly(ethylenimine). Their application

Product 2 was obtained in 79% yield after precipita-

in transition metal catalysis in batch and continuous tion from i-PrOH and was characterized with H, ¹³C,

processes is investigated and the results are discussed and ³¹P NMR and mass spectrometry. This model com-

and compared to shine light on the effect of the support pound was obtained in at least 97% purity as judged

on catalysis. It is, to the best of our knowledge, the ﬁrst from GC analysis. NMR spectroscopy does not give

time that dendrimers and hyperbranched polymers are conclusive information on the purity of the product

1

directly compared in this type of application.

because both in ¹H and ³¹P NMR spectrometry the sig-

nals of the desired and the byproduct overlap. In the

GC-MS spectrum it is possible to distinguish two peaks

corresponding to MW = 779 (desired product) and 608

(byproduct). The impurities are thus likely the result of

results And dIscussIon

Synthesis

The periphery of the commercially available poly- bis(phosphine)urea formation, as this corresponds to the

(propylenimine) dendrimer was functionalized with MW of 608 Da.

p-(Diphenylphosphino)benzylamine groups via urea The scope of the reaction was ﬁrst extended to the

linkage formation. First an appropriate ligand moiety smallest member of the poly(propylenimine) dendrimer

bearing a reactive amino function was prepared for family, bearing 4 end-groups, giving product 3 (Fig. 2).

further functionalization of the dendritic scaffolds.

Also in this case the product was obtained in high

1

13

p-(Diphenylphosphino)benzylamine 1 has been pre- yield (82%) and purity (98%) as judged from H, C,

pared (Scheme 1) following a slightly modiﬁed litera- and ³¹P NMR and mass spectrometry. The only impu-

ture procedure.¹⁰

rity present appears to be phosphine oxide, detectable

Scheme 1. Synthesis of p‑(Diphenylphosphino)benzylamine: Nucleophilic phosphinylation of ﬂuoro aromatic compounds.

Scheme 2. CDI-coupling of p-(diphenylphosphino)benzylamine via urea linkage to a polyamine scaffold.

Ribaudo et al. / Functionalized Dendritic Phosphorus Ligands

82

Fig. 2. Phosphine-functionalized 1st generation DAB-PPI-dendrimer.

Fig. 3. FAB-MS spectrum of the phosphine-functionalized 1st generation DAB-PPI-dendrimer.

31

both in P NMR and FAB MS. The mass spectrum of cedure was applied to the 5th generation of the

3 (Fig. 3) clearly shows the molecular peak at m/z = poly(propylenimine) dendrimer family, which has a

1586 Da, as well as its oxidized analogues. Most of the MW of 7168. The phosphine-functionalized compound

oxidation takes place during the mass analysis experi- 4 is depicted in Fig. 5.

ment.

Thanks to the well‑deﬁned structure of the parent

The peaks at 820, 1225, and 1268 Da are due to dendrimer, it is possible to dose accurately the amounts

fragmentations typically observed for this type of den- of reagents. The addition of exact stoichiometric

drimers (Fig. 4).¹²

amounts of CDI and phosphine with respect to the poly-

Encouraged by these results, the synthetic pro- amine makes the work‑up and puriﬁcation very simple.

Israel Journal of Chemistry

49

2009

83

Fig. 4. Molecular fragments observed in the FAB-MS for Phos-PPI-G1.0 (3) (Fig. 3).

Fig. 5. Schematic depiction of the phosphine-functionalized 5th generation DAB-PPI-dendrimer.

Fig. 6. Phosphine-functionalized model polyamine compounds mimicking structural elements of hyperbranched PEI.

Also in this case the product was obtained in high purity mer units. They both show a degree of branching of

1

(95%) as judged from H, ¹³C, and ³¹P NMR analysis. ~70% and a polydispersity of 1.3 and 2.5, respectively.

The impurities are a consequence of partial oxidation of

Two model “monomeric” compounds were ﬁrst pre-

the phosphorous moieties, similar to that observed for 3. pared, consisting respectively of primary/tertiary and

It was not possible to obtain satisfactory MALDI-TOF primary/secondary amino groups only. These were used

spectra as additional evidence for the formation of the as model compounds for the preparation of the function-

product, which is not uncommon for molecules of this alized hyperbranched poly(ethylenimine) and were also

type and size.¹²

used as reference compounds in catalysis. The products

A similar procedure for functionalization has been (along with the respective precursors) are displayed in

applied to two hyperbranched poly(ethylenimine) Fig. 6.

polymers, PEI-5 and PEI-25, with average molecular

The compounds Phos-TREN (5) and Phos-TETA (6)

weights (M_n) of, respectively, ca. 5000 and ca. 25,000, were characterized using ¹H, ¹³C, and ³¹P NMR and mass

corresponding to, respectively, ~180 and ~580 mono- spectrometry. According to these measurements 5 and 6

Ribaudo et al. / Functionalized Dendritic Phosphorus Ligands

84

were very pure (96% in both cases); the impurities de- we observed that 5% of the P ligands were oxidized after

rive again from partial oxidation (2%) of the P moieties the work-up procedure. The content of P in the function-

and bis(phosphine)urea formation (2%). The overall al hyperbranched polymers was determined by means

yields were 83% and 89%, respectively. of calibration versus a (simple) known phosphine mole-

The covalent functionalization of hyperbranched cule (tri-o-tolyl phosphine) that is structurally related to

poly(ethylenimine) has been achieved using a similar the compounds under investigation but whose chemical

synthetic procedure (Scheme 3).

shift in the ³¹P NMR is sufﬁciently separated from those

The less deﬁned structure of the hyperbranched poly- of the polymer, thus allowing a good comparison of the

mer requires the use of a slight excess (5–10% excess) two integral signals (Fig. 7).

of p-(diphenylphosphino)benzylamine and 1,1¢-carbon-

The functional Phos-PEI-5 and Phos-PEI-25 show a

yldiimidazole with respect to the calculated number phosphorus content of, respectively, 2.620 mmol/g and

of –NH and –NH₂groups (40 and 35, respectively, for 2.622 mmol/g; these values correspond to, respectively,

PEI-5, 200 and 180 for PEI-25). The excess of reactants 77 and 390 P atoms per molecule, which is in good

can be easily removed by precipitating the solid crude agreement with the number of reactive (that is, primary

product from boiling isopropanol (see Experimental and secondary) amino groups per molecule (75 and 380,

section for details). At room temperature the macro- respectively) reported by the producer. This conﬁrms

molecular products appear to be soluble only in a small that all primary and secondary amine groups of the poly-

number of solvents, which does not include i-PrOH. mer were converted into a urea-phosphine moiety.

The products 7 and 8 have been characterized with ¹H,

¹³C, and ³¹P NMR. It was not possible to obtain satisfac- Catalysis

tory MALDI-TOF spectra as additional evidence for

the formation of the products, which is not uncommon (2–8) have been studied as multivalent ligands in tran-

The phosphorus-functionalized dendritic molecules

for molecules of this type and size.¹³The coupling was sition metal catalysis. We focused our attention on the

1

quantitative and according to H NMR there were no Pd-catalyzed allylic substitution reactions (i.e., amina-

free amino (–NH₂or –NH) groups left. From ³¹P NMR tion and thiolation). Other researchers in our group have

Scheme 3. Schematic depiction of the synthetic procedure used for the covalent attachment of p-(diphenylphosphino)benzylam-

ine to hyperbranched PEI via urea linkage (CDI coupling).

Israel Journal of Chemistry

49

2009

85

Fig. 7. ³¹P NMR spectrum of Phos-PEI-5 7 (–5.7 ppm) vs. tri-o-tolyl phosphine (–18.9 ppm).

Scheme 4. General mechanism for palladium catalyzed allylic subsitution.

drimers. The larger dendritic ligands (4, 7, and 8) have

also been used in a continuous‑ﬂow membrane reactor.

The catalytic sites located (mainly) at the periphery of

the functional macromolecules are readily accessible to

the substrate and the dendritic supports are sufﬁciently

large to be retained in the reactor when a membrane with

molecular weight cutoff of 700 Da is used.

previously studied different phosphorus-functionalized

dendrimers for applications in transition metal catalysis,

in batch and continuous ﬂow systems.^14,15It was dem-

onstrated that the position of the catalytic sites and their

spatial separation are determined by the geometry of the

dendrimer, which is very important for the performance

of the catalyst. For the allylic substitution reaction it

was found that periphery-functionalized systems have

activities comparable to those of monomeric systems,

but application in a continuous‑ﬂow membrane reactor

showed that these catalyst systems can be very unstable.

In contrast, the core-functionalized dendrimers gave

slower but very stable catalysts. Moreover, using these

dendritic catalysts the selectivity was controlled by the

apolar microenvironment created by the dendrimer.

We now apply the functionalized dendritic molecules

(“perfect” dendrimers and hyperbranched polymers)

in transition metal catalysis (batch and continuous

ﬂow) and compare the results with the ones previously

obtained from core- and periphery-functionalized den-

Batch Processes; Pd-Catalyzed Allylic Substitution

Transition metal complexes with unsaturated hydro-

carbons that are π‑bonded to the metal have long been

the focus of organometallic chemistry. The transition

metal catalyzed allylic substitution reaction using a va-

riety of nucleophilic reagents is now a well-established

methodology in organic synthesis and is widely used to

construct complex organic molecules.¹⁶

The mechanism of the palladium-catalyzed allylic

substitution (Scheme 4) proceeds via a cationic (or neu-

tral) palladium π-allyl complex (b), which is formed

after oxidative addition of (a) to palladium. In the sec-

Ribaudo et al. / Functionalized Dendritic Phosphorus Ligands

86

ond step, the nucleophile attacks directly at the allylic might be the consequence of a late transition state²⁰

ligand to form the product (c), and a palladium(0) spe- (favored in presence of a weak nucleophile such as an

cies forms that re-enters the catalytic cycle.

amine) and the interaction of the counterion (^–O₂COMe

→ MeO^–) with the palladium center, which is known to

increase the rate of dynamic behavior of the allyl moi-

Allylic Amination

We studied palladium–crotyl complexes (crotyl = ety.²¹This all results in an isomerization of the syn com-

but-2-enyl) of the P-functionalized dendritic species as plex to the anti complex and vice versa. The resulting

catalysts in the allylic substitution reaction using mor- regioselectivity of the catalytic reaction is dependent on

pholine as the nucleophile and methyl crotyl carbonate the relative rates of syn–anti isomerization and nucleo-

as the substrate (Scheme 4).¹⁷In a typical experiment the philic attack²²(Scheme 6). In the case under investiga-

catalyst was formed by dissolving crotylpalladium chlo- tion the isomerization is probably fast enough to favor

ride dimer and dendritic ligand in CH₂Cl₂(unless other- the formation of (preferentially) the branched product.

wise stated). According to the ³¹P NMR spectrum of the

We anticipated that, due to its non-uniform nature,

complex Pd·7 (P/Pd ratio = 2.20) in CDCl₃(Fig. 8) two the catalytic activity of the metal complexes of the mac-

major species are formed. The peak around 30 ppm is romolecular ligand 7 (Phos-PEI-5) might be sensitive to

attributed to the mono-ligated species, whereas the one the metal loading (Table 2 and Fig. 9).

at ~22 ppm is attributed to the bidentate syn complex.

The curves displayed in the graph (Fig. 9) show the

After ~120 min of incubation time the reaction was conversion of substrate (crotyl methyl carbonate) into

started by the addition of a CH₂Cl₂solution containing the various products (allylamine isomers) as a function

crotyl methyl carbonate (reactant), morpholine (reac- of time; each curve corresponds to a different metal

tant and base) and n-decane (internal standard) to the loading. The catalytic activity of the “perfect” dendritic

catalyst solution. The reaction was monitored by gas ligand 4 is also displayed. Indeed very small changes

chromatography. in P/Pd ratio (from 1.99 to 2.10) resulted in substantial

In Tables 1 and 2 the results are displayed. The mo- changes in activity.

nomeric modular ligands 2-3-5-6 appear to form effec-

Ligand 4 provides palladium catalysts that show

tive palladium catalysts for the allylic amination of cro- activity comparable to the catalyst based on PPh₃(en-

tyl methyl carbonate. All reactions went to completion tries 1–3). Large changes in phosphorus/palladium ratio

within 1 h using 3 mol % of catalyst; this relatively high for ligand 4 gave much smaller differences of activity

catalyst loading was chosen to compare the results with and opposite than the hyperbranched polymer deriva-

previous work. Their activities are all comparable to the tive; an excess of phosphorus units leads to a decrease

activity of the catalyst formed by PPh₃. The activity of in activity. On the other hand, only a slight excess of

5 (Phos-TREN) increases upon increasing the P/Pd ra- ligand moieties (2.06 vs. 2) is necessary for the hyper-

tio; a similar trend has already been observed by other branched derivative 7 to equal the performance of the

authors.^14c,18The selectivity is in all cases in favor of the dendritic analogue (entries 4–7). This is attributed to

branched product. The proposed syn stereochemistry the non-perfect structure of the polymer compared to

of the Pd complex formed is not sufﬁcient to explain the dendrimer. Some P atoms are concealed in the bulk

the observed product distribution though.¹⁹In fact, this of the core of the macromolecule and are therefore not

Fig. 8. Part of the ³¹P NMR spectrum in chloroform-d (25 ºC) of the complex between Phos-PEI-5 (7) and [Pd(crotyl)Cl]₂.

Israel Journal of Chemistry

49

2009

87

Table 1. Pd-catalyzed allylic amination of crotyl methyl car- Table 2. Pd-catalyzed allylic amination of crotyl methyl car-

bonate and morpholine using “model” ligands P^a

bonate with morpholine using dendritic ligands 4-7-8^a

ligand

P / Pd

(molar

ratio)

TOF^b

product distribution

(%) branched/

linear trans/linear cis

entry

ligand

P / Pd

ratio

TOF^bproduct distribu-

tion (%) branched/

lineartrans/linearcis

PPh₃

2

3

2

286

229

277

180

299

387

65/30/5

64/32/4

64/31/5

60/35/5

58/38/4

64/30/6

1

2

3

4

5

6

7

8

9

10

Phos-G5.0 (4)

4

2.00

2.21

2.46

1.99

2.03

2.06

2.10

299

238

211

114

216

321

365

308

348

387

61 / 36 / 3

62 / 36 / 2

60 / 36 / 4

65 / 33 / 2

65 / 32 / 3

66 / 32 / 2

65 / 31 / 4

63 / 35 / 2

63 / 34 / 3

62 / 35 / 3

Phos-G0.0 (2)

Phos-G1.0 (3)

Phos-TREN (5)

Phos-TETA (6)

^aRoom temperature, solvent CH₂Cl₂, vol 3 mL, [substrate] =

0.11 M, [amine] = 0.12 M, [Pd] = 3 mM. ^bmol mol^–1h^–1; aver-

age turnover frequency measured after 5 min.

4

Phos-PEI-5 (7)

7

Phos-PEI-25 (8) 1.89

8

2.01

2.39

^aRoom temperature, solvent CH₂Cl₂, vol 3 mL, [substrate] =

110 mM, [amine] =120 mM, [Pd] = 3 mM. ^bmol mol^–1h^–1; av-

erage turnover frequency measured after 5 min.

accessible to form a bis-ligated palladium complex. It

is interesting to note that in absolute terms the hyper-

branched polymer-supported catalyst outperformed

the dendritic analogue. The activity is about 1.2 times

higher (entry 7 vs. entry 1). The back-folding phenom-

enon often observed with this class of compounds pro-

vides an additional and complementary explanation for

the opposite trends shown by dendrimer 4 and hyper-

branched polymer 7. In the ﬁrst case, due to the uniform

well‑deﬁned structure of dendrimer 4, the back-fold-

ing of the peripheral arms (in conditions of excess of

phosphorus vs. palladium) hampers the approaching of

substrate molecules towards the catalytic centers. In the

case of the hyperbranched polymer 7, the back-folding

is, on the other hand, beneﬁcial as it provides stabiliza-

tion of the metal centers, thus increasing the catalytic

activity. Comparing the catalytic activities of ligands 2,

3 (Table 1), and 4 (Table 2), a small positive dendritic

effect on the TOF can be noted, which is presumably a

consequence of the increasing bulk around the catalytic

metal center. A similar effect was observed by de Groot

et al. in the catalytic activity exhibited by different

phosphine-functionalized carbosilane dendrimers in the

Pd-catalyzed allylic amination.^14cA more pronounced

effect was reported by Gade and coworkers, who fol-

lowed the activity of different generations of Pyrphos-

functionalized PPI and PAMAM dendrimers.²³These

effects are often observed in palladium catalyzed allylic

substitutions, which are known to be particularly sensi-

tive to small changes in the chemical environment of the

active catalyst site.^16d,17a,24Experiments performed with

the larger member of the hyperbranched poly(ethylen-

imine) derivatives, i.e., Phos-PEI-25 (8), exhibited a

Fig. 9. Pd-catalyzed allylic amination of crotyl methyl carbonate with morpholine using dendritic ligands Phos-DAB-G5.0 (4)

and Phos-PEI-5 (7).

Ribaudo et al. / Functionalized Dendritic Phosphorus Ligands

88

Scheme 5. Pd-catalyzed allylic amination reaction of crotyl methyl carbonate with morpholine.

Scheme 6. Formation of regioisomers in the allylic substitution of (crotyl)Pd(bisphosphine) complexes.

Scheme 7. Allylic thiolation reaction of crotyl methyl carbonate with thiophenol.

similar trend as for macroligand 7 (entries 8–12), that is,

We were interested to see if our dendritic catalysts

an increase of activity upon increase of the P/Pd ratio. were also active in the allylic thiolation and therefore

The experiments were carried out in CH₂Cl₂as solvent studied the current system in this reaction (Scheme 7).

but the results suffered from inadequate reproducibility Indeed, palladium–crotyl complexes (crotyl = but-2-

(±30%), probably due to limited solubility of the cata- enyl) of the dendritic species were active catalysts in

lyst system in this solvent. The reactions were also per- the allylic substitution reaction performed using thio-

formed in CH₃CN and THF, achieving clear solutions, phenol as the nucleophile and crotyl methyl carbonate

but they showed low conversion values, possibly as a as the substrate. In a typical experiment the catalyst was

result of the coordinating properties of the aforemen- formed by dissolving crotylpalladium chloride dimer

tioned solvents.

and dendritic ligand in CH₂Cl₂(unless otherwise stated).

After an incubation time of ~ 120 min, the reaction was

started by addition of a CH₂Cl₂solution containing cro-

Allylic Thiolation

Transition metal-catalyzed synthetic reactions using tyl methyl carbonate (reactant), thiophenol (reactant),

sulfur-containing compounds are not straightforward, 4-methylmorpholine = N-methylmorpholine (NMM)

since these compounds can act as catalyst poisons be- (auxiliary base) and n-decane (internal standard) to the

cause of the strong coordinating properties^25–27of sulfur. catalyst solution. The reaction was monitored by gas

Recently, a palladium-catalyzed S-allylation of aro- chromatography.

matic and heteroaromatic thiols with allylic carbonates

All reactions went to completion within 1 h except

has been reported.^28,29Allylic rearrangements consistent the experiment carried out with macroligand 7 applied

with a π‑allyl palladium intermediate are observed. Re- at low P/M ratio; in Table 3 the turnover frequency after

gioisomeric allylic carbonates gave identical mixtures 5 min is displayed to compare activities.

of regio‑ and stereoisomeric sulﬁdes in which the sub-

The ligands under study appear to form effective

stitution occurred at the less hindered allylic terminus of palladium catalysts for the allylic thiolation of crotyl

the π‑allyl system.

methyl carbonate, although the reaction rates are lower

Israel Journal of Chemistry

49

2009

89

Table 3. Pd-catalyzed allylic thiolation of crotyl methyl car-

bonate with thiophenol using dendritic ligands 2-4-7^a

some P atoms are concealed in the core of the macro-

molecule, which are less accessible for formation of

bis(phosphine)palladium complexes.

ligand

P/Pd

(molar

ratio)

TOF P^bproduct distribution (%)

branched/linear trans/

linear cis

Continuous Processes

Reactor setup

PPhB_3B

Phos-G0.0 (2)

Phos-G5.0 (4)

Phos-PEI-5 (7) 1.99

7

^aRoom temperature, solvent CH₂Cl₂, vol 3 mL, [substrate] =

0.11 M, [thiol] = 0.12 M, [base] = 0.02 M, [Pd] = 3 mM. ^bmol

mol^–1h^–1; average turnover frequency measured after 5 min.

2

198

224

216

75

154

233

67/30/3

36/60/4

47/50/3

40/55/5

42/53/5

38/57/5

A continuous‑ﬂow membrane reactor (CFMR) pre-

viously developed in our group was used for various

catalytic reactions (Fig. 10a). The reactor consists of a

stainless steel autoclave in which the catalyst solution

can be injected via a two-way valve. The membrane is

located at the bottom of the autoclave and kept in place

by a Viton O‑ring. The Koch/SelRO MPF‑50 nanoﬁltra-

tion membrane (MW cutoff = 700 Da)³⁰was chosen for

the ﬁltration experiments since it previously gave good

results with similar dendritic ligands in transition metal-

catalyzed reactions.¹⁴For similar reasons CH₂Cl₂was

used to prepare all solutions because it proved to be the

best solvent so far.¹⁴A vessel containing the substrate

solution (usually under inert atmosphere) is connected to

the membrane reactor via an HPLC-pump that pushes the

substrate solution into the reactor and through the mem-

brane. The product stream is collected and samples are

taken at regular time intervals; these are analyzed by GC

to obtain the conversion in the product phase as a func-

tion of time.

2.03

2.06

than those of the allylic amination reaction. Their activi-

ties are all comparable with the activity of the simple

PPh₃-based catalyst, while the selectivity is in this case

in favor of the linear isomers, unlike for the allylic ami-

nation. The trends in catalytic activity and selectivity are

consistent with the “late” nature of the transition state,

which is more pronounced for S than for N.^20,22The

sulfur nucleophile, which can be considered a “soft”

nucleophile, attacks preferentially at the less substituted

termini of the π‑allyl intermediate.^28a,b,P

In analogy with the allylic amination, very small

changes in P/Pd ratio (from 1.99 to 2.06) result in sub-

stantial changes in TOF. Again, it may be inferred that a

slight excess of ligand moieties is necessary for the hy-

perbranched derivative to approach the performances of

the dendritic analogue, probably due to the non-perfect

structure of the polymer compared to the dendrimer;

Commonly, dendritic transition metal catalysts ap-

plied in a CFMR can suffer from two forms of leaching:

depletion of the dendritic catalyst through the mem-

brane and metal dissociation from the dendrimer result-

ing in further leaching of the unsupported metal through

the membrane. In Fig. 10b the theoretical retention of

Fig. 10. Schematic presentation of a membrane reactor (left, a) and theoretical relative concentrations (C_r) of the dendritic spe-

cies vs. the substrate ﬂow (expressed in number of reactor volumes pumped through N_r) calculated for various retention coef-

ﬁcients (right, b).

Ribaudo et al. / Functionalized Dendritic Phosphorus Ligands

90

Fig. 11. Product yield in the allylic amination of crotyl methyl carbonate with morpholine in the CFMR as a function of the sub-

strate ﬂow (expressed as no. of reactor volumes) using different dendritic ligands (4), (7), and (8). {Reaction conditions: [sub-

strate] = 0.11 M, [amine] = 0.12 M, [n-decane] = 0.10 M, [Pd] = 5 mM, 6 mL CH₂Cl₂, room temperature, ﬂow rate = 9 mL h^–1

(150 µL min^–1)}.

catalysts are plotted for dendritic systems with vari-

ous retention coefﬁcients. For example, if a dendritic

catalyst has a retention factor of 0.95, only 25% of the

catalyst would remain in the reactor after the reactor had

been ﬂushed with 30 times its volume. For practical ap-

plications, the overall retention of the dendritic catalyst

must be very high (typically, >99.9%) to maintain the

material in a CFMR for long reaction times.

formances shown by the three ligands under investi-

gation displays the differences between the various

supports.

The recycling with macroligand 4 was performed

using Pd catalyst (5 mol % catalyst vs. substrate, P/Pd

ratio = 2). The reaction started immediately after addi-

tion of the catalyst solution and reached its maximum

conversion (~90%) after approximately two cycles.

The yield gradually started decreasing during the next

thirty-eight cycles, after which the reaction was stopped

because all substrate solution was consumed. The ob-

served decrease in catalyst activity was ascribed to loss

of the palladium compound from the reactor; this was

also conﬁrmed by the slight yellow color of the prod-

uct stream^{3,14a,c,15,31}(a similar though far more dramatic

phenomenon was also observed with dendritic Ni cata-

lysts³²). Loss of the dendritic ligand as a possible reason

for the drop in activity was excluded by analysis of the

Allylic amination

The macromolecular ligands Phos-DAB-G5.0 (4),

Phos-PEI-5 (7) and Phos-PEI-25 (8) were applied in a

CFMR for the allylic amination of crotyl methyl car-

bonate. A CH₂Cl₂solution of the ligand under study

(4, 7, or 8) and crotylpalladium chloride dimer was

stirred at room temperature for 3 h and then 3 mL was

injected into the membrane reactor (vol 6 mL). At the

same time the reactor was fed with a solution of crotyl

methyl carbonate, morpholine, and n-decane (internal

standard). All experiments performed resulted in a pres-

sure buildup in the membrane (up to 10–11 bar), as a

consequence of the production of CO₂as one of the

reaction byproducts. The product stream of the reaction

was sampled at regular intervals of time and analyzed

by means of GC; the conversion values are plotted as

a function of the number of cycles performed by the

reactor. The curves displayed show the product yield as

a function of the substrate ﬂow, indicated as number of

reactor volumes:

31

reactor content at the end of the reaction. P NMR of

the solution in the reactor, vs. tri-o-tolylphosphine as a

standard, showed that the P-functionalized macromol-

ecule was completely retained in the reactor (within the

experimental error), as expected on the basis of its size.

Interestingly, the stereoselectivity observed during the

continuous recycling experiment is similar to that of the

batch processes, in favor of the branched product. When

applying ligand 7, the Pd catalyst was used in 5% molar

ratio vs. substrate and a P/Pd ratio of 2.3 was applied

(slightly higher than the “optimum” found in the corre-

sponding batch reaction). The reaction starts immediate-

ly after addition of the catalyst solution with very high

conversion (~94%). The yield gradually decreased dur-

#(reactor volumes) = (ﬂow rate ´ time)/(reactor volume)

A comparative view (Fig. 11) of the catalytic per-

Israel Journal of Chemistry

49

2009

91

ing the next thirty-two hours (forty-eight cycles). Also might be responsible for the fast loss of activity shown

in this case the observed decrease in catalyst activity by Phos-PEI-25. It has been shown that the globular and,

was ascribed to loss of palladium from the reactor^3,14a,c,31perhaps more important, shape-persistent structures of

(the outgoing solution was slightly yellow). Similarly the dendrimers are fundamental for effective applica-

31

to the case of the perfect dendritic ligand 4, P NMR tion in membrane ﬁltration recycling; a high degree of

of the solution in the reactor vs. tri-o-tolylphosphine rigidity in the backbone of macromolecular complexes

showed that the P-functionalized macromolecule was leads indeed to more efﬁcient retentions of multimetal-

completely retained in the reactor. The observed selec- lic materials by nanoﬁltration membranes.³³A number

tivity is again similar to that of the corresponding batch of experimental^34,35and theoretical³⁸approaches show

process (predominantly branched product). In spite of that dendrimers and hyperbranched polymers in good

the difﬁculties encountered during the batch experi- solvent conditions³⁷are best described as ﬂexible mac-

ments (solubility), we chose to investigate the perfor- romolecular aggregates^[38]with their maximum density

mance of ligand 8 in CFMR. The Pd catalyst was used in the center of the molecule (dense-core structure)³⁹and

in 5% molar ratio vs. substrate and P/Pd ratio of 2.39 ﬂuctuating monomer groups.⁴⁰The higher ﬂexibility of

was applied (slightly higher than the “optimum” found Phos-PEI-25 compared to Phos-PEI-5 could contribute

in the corresponding batch reaction). The curve shows to the formation of less stable catalysts and therefore

that the reaction starts immediately after addition of the cause faster loss of activity of the system when applied

catalyst solution with very high conversion (~98%). The in the CFMR.

yield immediately decreased with a steep slope during

the next fourteen hours (twenty-two reactor cycles).

The observed decrease in catalyst activity was again

Allylic thiolation

The macromolecular ligand Phos-PEI-5 (7) was ap-

attributed to loss of palladium from the reactor;^3,14a,c,31plied in a continuous‑ﬂow membrane reactor for the

this was conﬁrmed by the yellow color of the product allylic thiolation of crotyl methyl carbonate. A CH₂Cl₂

stream and total retention of the dendritic ligand within solution of the ligand and crotylpalladium chloride di-

the nano-membrane reactor.

mer was mixed at room temperature and then injected

The structure and stability of the metal complexes of into the membrane reactor. Subsequently the reactor

the phosphine-functionalized molecules, which is cru- was fed with a solution of crotyl methyl carbonate (re-

cial for the application in a continuous process, appear actant), thiophenol (reactant), NMM (auxiliary base)

to be very sensitive to changes in the dendritic struc- and n-decane (internal standard). The setup of the

ture. This was also previously found for carbosilane- recycling system is done in a similar way as for the

based dendrimers. When using dendrimer ligands with allylic amination. The product stream of the reaction

Si(CH₂PPh₂)₂end-groups, the yield for allylic substitu- was sampled at regular intervals of time and analyzed

tion reactions dropped rapidly during continuous ﬂow by means of GC; the conversion values are plotted as

experiments, whereas a stable catalyst was obtained in a function of the number of cycles performed by the

a CFMR when dendrimer ligand with SiCH₂CH₂PPh₂reactor in Fig. 12.

end-groups was used.^14c,15

The reaction started immediately after addition of

Importantly, the behavior displayed by the “perfect’ the catalyst solution with high conversion (~83%) and

dendritic ligand Phos-G5.0 (4) and the smaller hyper- reached its maximum value (~96%) after approximately

branched Phos-PEI-5 (7) is quite similar. This was sup- three cycles. The conversion then decreased during the

ported by the small amount of black Pd metal deposited next twelve cycles. It is reasonable to assume that, in

on the membrane in both cases. The curves of the men- analogy with the allylic amination reaction, structural

tioned ligands (represented in the diagram by circles and features lead to a drop in catalytic activity. In this exper-

squares, respectively) diverge in the initial part but then iment the product stream appeared much more intensely

assume a very similar declining trend; likely the Pd- colored (yellow) than found for the allylic amination

complexes formed (the active catalysts) are very similar reactions; this phenomenon is probably a consequence

in nature. The clearly faster loss in activity shown by the of the strong afﬁnity of Pd metal centers for S ligands

Phos-PEI-25 (lig. 8) compared to the smaller analogue (reactant and products, in this case) that would eventu-

Phos-PEI-5 (lig. 7) can be attributed to the lower stabil- ally wash out the metal catalyst.^26,31P NMR of the solu-

ity of the Pd complexes formed, as can be deduced from tion in the reactor vs. tri-o-tolyl phosphine showed that

the more intense yellow color of the outgoing stream and the P-functionalized macromolecule was completely re-

the distinctly larger amount of Pd metal accumulated on tained in the reactor. The observed selectivity is similar

the membrane. The aforementioned solubility issues to that of the corresponding batch process, i.e., the linear

and the structural characteristics of the macromolecule isomers prevail.

Ribaudo et al. / Functionalized Dendritic Phosphorus Ligands

92

Fig. 12. Product yield in the allylic thiolation of crotyl methyl carbonate with thiophenol in the CFMR as a function of the sub-

strate ﬂow (expressed as no. of cycles) using dendritic ligand 7. {[substrate] = 0.11 M, [thiol] = 0.12 M, [base] = 0.02 M, [n-

decane] = 0.12 M, [Pd] = 5 mM, P/Pd = 2.31, 6 mL CH₂Cl₂, room temperature, ﬂow rate = 6 mL h^–1(100 µL min^–1)}.

conclusIons

experImentAl sectIon

General Remarks

In this paper we have described the facile and straight-

forward covalent functionalization of commercially

available dendritic poly(propylenimine) and hyper-

branched poly(ethylenimine) with P-containing func-

tional groups. The P-functionalized macromolecules

have been applied as multivalent ligands in the Pd-cata-

lyzed allylic substitution reactions (batch and continu-

ous process) using either morpholine or thiophenol as

nucleophile. Palladium complexes of all described mol-

ecules are active in allylic substitution reactions. The

PEI-functionalized polymers appear more sensitive to

small changes in the P/Pd ratio than the PPI analogues,

but form catalysts that are more active. In the allylic am-

ination, Pd complexes of Phos-PEI-5 show selectivities

comparable with PPh₃, whereas in the allylic thiolation,

the use of Phos-PEI-5 reverses the selectivity (predomi-

nantly linear allylic sulﬁdes). When the dendritic li-

gands are used in a continuous‑ﬂow process, the macro-

molecules are completely retained by the nanoﬁltration

membrane, while the catalytic activity decreases with

time because of palladium depletion. Palladium leach-

ing is not associated with the nature of the dendrimers

or macromolecules, but it is an intrinsic problem of pal-

ladium-catalyzed coupling reactions. Palladium deple-

tion is more severe for the allylic thiolation, probably

because of the stronger afﬁnity of sulfur for palladium,

facilitating palladium leaching. Unexpectedly, the larger

PEI-25 derivative loses activity more quickly than the

smaller Phos-PEI-5 ligand, showing that this is not a

suitable support for this reaction. In contrast, Phos-PEI-

5 shows a very similar behavior to the perfect DAB-den-

drimer derivative when applied in the CFMR. Indeed,

the cheap PEI polymer was successfully used as catalyst

support and outperformed the PPI dendrimer.

Allreactionswerecarriedoutinﬂamed‑driedglasswareand

under a dry nitrogen atmosphere using Schlenck techniques

unless mentioned otherwise. Solvents were dried and distilled

under nitrogen; Et₂O and THF from sodium/benzophenone,

hexane from sodium/benzophenone/triglyme, CH₂Cl₂and

EtOAc from CaH₂. Dendritic poly(propylenimines) (DAB

dendrimers, DSM products) were purchased from Aldrich

Chemical Company, hyperbranched poly(ethylenimine) PEI-

5 and PEI-25 were purchased from HyperPolymers GmbH,

other chemicals were purchased from Acros Organics or

Aldrich Chemical Company and used without further puri-

ﬁcation unless otherwise stated. The N-methyl morpholine

and morpholine were distilled prior to use, respectively, from

KOH and sodium. Crotyl methyl carbonate was prepared us-

ing a slightly modiﬁed literature procedure.⁴¹Silica 60 (SDS

Chromagel, 70–200 µm) was used for column chromatogra-

1

phy. H, ¹³C{¹H} and ³¹P{¹H} NMR spectra were recorded

using the following spectrometers: Varian Mercury 300 (¹H

300 MHz, ³¹P 121 MHz, and ¹³C 125 MHz) and Varian Inova

500 (¹³C 125 MHz). The spectra were recorded in CDCl_3Bun-

less stated otherwise and the chemical shifts are given in ppm

versus TMS (¹H, ¹³C) or 85% H₃PO₄(³¹P). FAB MS experi-

ments were performed on a Joel JMS SX/SX 102A four sector

mass spectrometer coupled to a Joel MS-MP7000 data system,

using 3-nitro benzyl alcohol as matrix. GC analyzes were

performed on an Interscience HRGC Mega 2 (J&W Scientiﬁc

DB-1 30 m ´ 0.32 mm ´ 3.0 µm). All catalytic experiments

were performed at least in duplicate.

p‑(diphenylphosphino)benzylamine (1).¹⁰A 250 mL

ﬂame‑dried three‑necked round bottom ﬂask equipped with a

nitrogen inlet and a reﬂux condenser was charged with 5 mL

of 4‑ﬂuorobenzylamine (43.95 mmol) and 88 mL of a 0.5 M

solution of KPPh₂in THF (Aldrich). After heating the reaction

mixture at reﬂux for 20 h the possible unreacted phosphido

anion was quenched with 5 mL of MeOH, then the solvent

Israel Journal of Chemistry

49

2009

93

was evaporated under reduced pressure, redissolved in CH₂Cl₂NCH₂CH₂CH₂CH₂N) 1.64 (m, 8 H, NCH₂CH₂CH₂N), 2.29–

(75 mL), and washed with degassed water (2 ´ 40 mL). The 2.34 (m, 12 H, NCH₂CH₂CH₂NHCO + NCH₂CH₂CH₂CH₂N),

organic phase was dried over Na₂SO₄, ﬁltered, and the solvent 3.10 (m, 8 H, NCH₂CH₂CH₂NHCO), 4.17 (d, 8 H,

removed under reduced pressure. The crude product was pu- ArCH₂NHCO), 6.10–6.50 (br., 8 H, NHCONH), 7.13–7.49

riﬁed using two methods different from the one reported in (m, 28 H, ArH); ¹³C{¹H} NMR (75 MHz, CDCl₃) δ 26.0,

the reference. Method A: the crude product was puriﬁed by 26.6, 39.9, 46.1, 51.1, 53.9, 127.5, 128.6, 128.8, 133.6, 134.1,

column chromatography over silica using CHCl₃/MeOH (9:1) 136.2, 137.3, 140.1, 158.2; ³¹P{¹H} NMR (121 MHz, CDCl₃)

as an eluent (R.F. = 0.3). Method B: the crude product was δ –5.27; FAB-MS [M]⁺1586.

puriﬁed by reprecipitation from boiling hexane. Total yield:

phos‑dAB‑Am‑64 (4). A 100 mL ﬂame‑dried Schlenck

1

8.96 g (70%), white product. H NMR (300 MHz, CDCl₃) δ ﬂask equipped with a 50 mL dropping funnel was charged

7.4–7.2 (m, 14 H, ArH), 3.8 (s, 2 H, CH₂), 2.2 (s (br), 2 H, with a solution of CDI, 0.476 g = 2.94 mmol, and NMM,

NH₂); ¹³C{¹H} NMR (75 MHz, CDCl₃) δ 46.3, 127.9, 128.9, 0.1 mL = 0.092 g = 0.91 mmol, ~1/3 equiv) in 5 mL of CHCl₃

129.3, 134.0, 134.7, 136.0, 137.8, 143.9; ³¹P{¹H} NMR (121 (fractionally distilled from CaCl₂). A solution of p-(Diphen

MHz, CDCl₃) δ –5.63 (s).

ylphosphino)benzylamine (0.778 g = 2.67 mmol) in CHCl₃

phos‑mA‑Am‑2 (2). A 50 mL ﬂame‑dried Schlenck (15 mL) was transferred into the dropping funnel and subse-

ﬂask equipped with a 25 mL dropping funnel was quently slowly dropped into the ﬂask and allowed to react for

charged with a solution of 1,1¢-carbonyl diimidazole ~15 min. A solution of DAB-Am-64 (0.272 g = 37.94 µmol,

(CDI, 0.487 g = 3.00 mmol) and NMM, 0.11 mL = 0.101 g = ~1/64 equiv) in THF (20 mL) was then added in one por-

1.00 mmol, ~1/3 equiv) in 5 mL of CH₂Cl₂. A solution of tion and left stirring for 2 ½ days at room temperature. The

p-(Diphenylphosphino)benzylamine (0.795 g = 2.73 mmol) solvent was evaporated and the crude product redissolved

in CH₂Cl₂(17 mL) was transferred into the dropping funnel in CH₂Cl₂(40 mL). The solution was then washed with de-

and subsequently slowly dropped into the ﬂask and allowed gassed water (2 ´ 20 mL) and dried over Na₂SO₄. Filtration

to react for ~15 min. A solution of 3,3¢-diamino-N-methyldi- and evaporation of the solvent under reduced pressure left

propylamine (0.200 mL = 0.180 g = 1.24 mmol, ~1/2 equiv) a white solid that was puriﬁed by reprecipitation from boil-

in CH₂Cl₂(3 mL) was then added in one portion and left ing i‑PrOH (40 mL). The ﬁltered product was crushed and

stirring overnight at room temperature. The reaction mixture further washed with pentane and/or Et₂O to remove traces

was washed with degassed water (2 ´ 20 mL) and then dried of solvent and give 0.826 g of white product (yield 79% vs.

1

over Na₂SO₄. Filtration and evaporation of the solvent under starting material DAB-Am-64). H NMR (300 MHz, CDCl₃)

reduced pressure left a white solid that was puriﬁed by re- δ 1.39 (m, 4 H, NCH₂CH₂CH₂CH₂N), 1.55 (br., 248 H,

precipitation from boiling i‑PrOH. The ﬁltered product was NCH₂CH₂CH₂N),2.11–2.48(br.,372H,NCH₂CH₂CH₂NHCO +

further washed with pentane and/or Et₂O to remove traces of NCH₂CH₂CH₂N + NCH₂CH₂CH₂CH₂N), 2.98–3.18 (br.,

solvent and give 0.764 g of white product (yield 79% vs. start- 128 H, NCH₂CH₂CH₂NHCO), 3.97–4.36 (br., 128 H,

ing material MA-Am-2). ¹H NMR (300 MHz, CDCl₃) δ 1.57 ArCH₂NHCO), 6.08–6.57 (br., 128 H, NHCONH), 6.91–7.69

(t, 4 H, NCH₂CH₂CH₂N), 2.22 (s, 3 H, CH₃N), 2.34 (t, 4 H, (m, 896 H, ArH); ¹³C{¹H} NMR (125 MHz, CDCl₃) δ 24.4,

NCH₂CH₂CH₂NHCO), 3.12 (m, 4 H, NCH₂CH₂CH₂NHCO), 27.7, 37.5, 45.7, 51.1, 51.7, 127.5–140.1, 158.2; ³¹P{¹H}

4.16 (d, 4 H, ArCH₂NHCO), 6.08–6.47 (br., 4 H, NHCONH), NMR (121 MHz, CDCl₃) δ –5.32 (br.).

7.11–7.45 (m, 28 H, ArH); ¹³C{¹H} NMR (75 MHz, CDCl₃) δ

phos‑tren (5). A 50 mL ﬂame‑dried Schlenck ﬂask

27.0, 38.8, 43.6, 44.3, 52.7, 127.5, 128.6, 128.8, 133.6, 134.1, equipped with a 25 mL dropping funnel was charged with a

136.2, 137.3, 140.1, 158.2; ³¹P{¹H} NMR (121 MHz, CDCl₃) solution of CDI, 0.476 g = 2.94 mmol, and NMM, 0.1 mL =

δ –5.41; GC-MS [M]⁺779.

0.092 g = 0.091 mmol, ~1/3 equiv) in 5 mL of CH₂Cl₂. A so-

phos‑dAB‑Am‑4 (3). A 100 mL ﬂame‑dried Schlenck lution of p-(Diphenylphosphino)benzylamine (0.778 g = 2.67

ﬂask equipped with a 50 mL dropping funnel was charged mmol) in CH₂Cl₂(20 mL) was transferred into the dropping

with a solution of CDI, 0.952 g = 5.87 mmol, and NMM, funnel and subsequently slowly dropped into the ﬂask and

0.2 mL = 0.184 g = 1.82 mmol, ~1/3 equiv) in 8 mL of CH₂Cl₂. allowed to react for ~10 min. A solution of TREN (tris(2-

A solution of p-(Diphenylphosphino)benzylamine (1.555 g = aminoethyl)amine) (0.120 mL = 0.120 g = 0.822 mmol, ~1/3

5.34 mmol) in CH₂Cl₂(20 mL) was transferred into the drop- equiv) in THF (3 mL) was then added in one portion and left

ping funnel and subsequently slowly dropped into the ﬂask stirring overnight at room temperature. The reaction mixture

and allowed to react for ~15 min. A solution of DAB-Am-4 was washed with degassed water (2 ´ 15 mL) and then dried

(0.400 mL = 0.384 g = 1.21 mmol, ~1/4 equiv) in CH₂Cl₂over Na₂SO₄. Filtration and evaporation of the solvent un-

(7 mL) was then added in one portion and left stirring over- der reduced pressure left a white solid that was puriﬁed by

night at room temperature. The reaction mixture was washed reprecipitation from boiling Et₂O. The ﬁltered product was

with degassed water (2 ´ 20 mL) and then dried over Na₂SO₄. further washed with pentane to remove traces of solvent and

Filtration and evaporation of the solvent under reduced pres- give 0.831 g of white product (yield 92% vs. starting mate-

1

sure left a white solid that was puriﬁed by reprecipitation rial TREN). H NMR (300 MHz, CDCl₃) δ 2.57 (t., 6 H,

from boiling i‑PrOH. The ﬁltered product was further washed NHC(O)NHCH₂CH₂N), 3.29 (t., 6 H, NH(O)CNHCH₂CH₂N),

with pentane and/or Et₂O to remove traces of solvent and 4.32 (s., 6 H, ArCH₂NHCO), 6.05–6.61 (br., 6 H, NHCONH),

give 1.58 g of white product (yield 82% vs. starting material 7.24–7.63 (m., 42 H, ArH); ¹³C{¹H} NMR (75 MHz, CDCl₃)

1

DAB-Am-4). H NMR (300 MHz, CDCl₃) δ 1.39 (m, 4 H, δ 40.7, 43.2, 51.3, 127.3, 128.8, 129.2, 133.3, 134.5, 136.7,

Ribaudo et al. / Functionalized Dendritic Phosphorus Ligands

94

138.0, 140.1, 157.3; ³¹P{¹H} NMR (121 MHz, CDCl₃) δ and left stirring for 5 dd at room temperature. The solvent was

–5.47; FAB-MS [M]⁺1098.

evaporated and the crude product puriﬁed by reprecipitation

phos‑tetA (6). A 50 mL ﬂame‑dried Schlenck ﬂask from boiling i-PrOH (60 mL); the solid was isolated and the

equipped with a 25 mL dropping funnel was charged with last procedure repeated 2 more times to remove all impurities.

a solution of CDI, 0.479 g = 2.95 mmol, and NMM, 0.1 The ﬁltered product was crushed and further washed with pen-

mL = 0.092 g = 0.091 mmol, ~1/3 equiv) in 5 mL of CH₂Cl₂. tane to remove traces of solvent and give 0.751 g of off-white

1

A solution of p-(Diphenylphosphino)benzylamine (0.782 product (yield 69% vs. starting material PEI-25). H NMR

g = 2.69 mmol) in CH₂Cl₂(20 mL) was transferred into the (300 MHz, CDCl₃) δ 2.29–2.73 (br., CH₂NCH₂CH₂NCH₂),

dropping funnel and subsequently slowly dropped into the 2.98–3.42 (br., CH₂N(CO)CH₂CH₂NHCO), 3.91–4.37 (br.,

ﬂask and allowed to react for ~10 min. A solution of TETA ArCH₂NHCO), 5.97–6.52 (br., NHCONH), 6.95–7.75 (m,

(triethylenetetramine) (0.100 mL = 0.098 g = 0.672 mmol, ArH); ¹³C{¹H} NMR (125 MHz, CDCl₃) δ 38.7, 46.2, 50.3,

~1/4 equiv) in CH₂Cl₂(3 mL) was then added in one portion 51.2, 127.3, 128.3, 129.1, 133.2, 134.8, 136.5, 137.7, 140.2,

and left stirring overnight at room temperature. The reac- 158.4; ³¹P{¹H} NMR (121 MHz, CDCl₃) δ –5.32 (br.).

tion mixture was washed with degassed water (2 ´ 15 mL)

crotyl methyl carbonate. A solution of 50.0 mL

and then dried over Na₂SO₄. Filtration and evaporation of (42.25 g = 0.586 mol) crotyl alcohol (97% trans) and 71.0 mL

the solvent under reduced pressure left a white solid that (69.44 g = 0.878 mol) pyridine in 60 mL CH₂Cl₂was cooled

was puriﬁed by reprecipitation from boiling Et₂O. The ﬁl- to 0 ºC with an ice/water bath, after which 54.0 mL (66.04 g =

tered product was further washed with pentane to remove 0.699 mol) methyl chloroformate was added dropwise. The re-

traces of solvent and give 0.884 g of white product (yield action mixture was left to warm to room temperature and was-

93% vs. starting material TETA). ¹H NMR (300 MHz, stirred overnight, after which it was poured into 500 mL 10%

CDCl₃) δ 3.35 (br. m., 12 H, NH(O)CNHCH₂CH₂NC(O)NH + NH₄Cl solution. The organic layer was separated, washed with

NHC(O)NCH₂CH₂NC(O)NH), 4.27 (s., 8 H, ArCH₂NHCO), 250 mL brine, dried over Na₂SO₄, and concentrated in vacuo.

6.09–6.47 (br., 6 H, NCONH + NHCONH), 7.21–7.59 (m., 56 The crude product was puriﬁed by distillation under reduced

H, ArH); ¹³C{¹H} NMR (75 MHz, CDCl₃) δ 39.2, 43.5, 47.2, pressure to give 63.29 g of colorless liquid (yield = 83%). ¹H

50.5, 126.9, 127.9, 128.3, 132.9, 133.7, 135.8, 137.1, 141.2, NMR (300 MHz, CDCl₃) δ 1.71 (d, CH₃CH = CHCH₂, 3 H),

159.0; ³¹P{¹H} NMR (121 MHz, CDCl₃) δ –5.61; FAB-MS 3.67 (s, OCH₃, 3 H), 4.67 (d, CH = CHCH₂O, 2 H), 5.63 (m,

[M]⁺1416.

CH₃CH = CHCH₂, 1 H), 5.73 (m, CH₃CH = CHCH₂, 1 H);

phos‑peI‑5 (7). A 100 mL ﬂame‑dried Schlenck ﬂask ¹³C{¹H} NMR (75 MHz, CDCl₃) δ 18.4, 55.4, 69.8, 125.3,

equipped with a 50 mL dropping funnel was charged with a 131.5, 155.7.

solution of CDI, 0.773 g = 4.77 mmol, and NMM, 0.15 mL =

0.14 g = 1.36 mmol, ~1/3 equiv) in 8 mL of THF. A solution

of p-(Diphenylphosphino)benzylamine (1.389 g = 4.77 mmol)

in THF (22 mL) was transferred into the dropping funnel and

subsequently slowly dropped into the ﬂask and allowed to

react for ~ 20 min. A solution of PEI-5 (0.298 g = 59.6 µmol,

~1/80 equiv) in THF (25 mL) was then added in one portion

and left stirring for 4 dd at room temperature. The solvent was

evaporated and the crude product puriﬁed by reprecipitation

from boiling i-PrOH (60 mL); the solid was isolated and the

last procedure repeated 2 more times to remove all impurities.

The ﬁltered product was crushed and further washed with pen-

Batchwise Allylic Amination Catalysis

A solution of ligand (18 μmol of “P” units, P/Pd = 2) and

[Pd(crotyl)Cl]₂(1.773 mg = 9 µmol of “Pd”) in CH₂Cl₂(2 mL)

was stirred for 2 h at R.T. Substrate solution was prepared by

dissolving 0.0716 g crotyl methyl carbonate (0.55 mmol),

0.0523 g morpholine (0.6 mmol) and 0.0712 g n-decane (0.5

mmol) in 5 mL CH₂Cl₂. Addition of 1 mL of substrate solution

started the catalysis. Samples of 50 µL were taken at t = 5, 10,

15, 20, 25, 30, 40, 50, 60 min, diluted in 3 mL Et₂O, ﬁltered

over Celite, and analyzed by GC.

Batchwise Allylic Thiolation Catalysis

A solution of ligand (18 µmol of “P” units, L/M = 2) and

[Pd(crotyl)Cl]₂(1.773 mg = 9 µmol of “Pd”) in CH₂Cl₂(2

mL) was stirred for 2 h at R.T. Substrate solution was prepared

by dissolving 0.0716 g crotyl methyl carbonate (0.55 mmol),

0.0661 g thiophenol (0.6 mmol), 0.0101 g NMM (0.1 mmol)

and 0.0712 g n-decane (0.5 mmol) in 5 mL CH₂Cl₂. Addition

tane to remove traces of solvent and give 1.081 g of off-white

1

product (yield 63% vs. starting material PEI-5). H NMR

(300 MHz, CDCl₃) δ 2.36–2.78 (br., CH₂NCH₂CH₂NCH₂),

2.96–3.47 (br., CH₂N(CO)CH₂CH₂NHCO), 3.87–4.49 (br.,

ArCH₂NHCO), 6.06–6.62 (br., NHCONH), 6.97–7.72 (m,

ArH); ¹³C{¹H} NMR (125 MHz, CDCl₃) δ 38.5, 45.7, 50.5,

51.7, 127.5, 128.6, 128.8, 133.6, 134.1, 136.2, 137.3, 140.1,

of 1 mL of substrate solution started the catalysis. Samples of

50 µL were taken at t = 5, 10, 15, 20, 25, 30, 40, 50, 60 min, di-

158.2; ³¹P{¹H} NMR (121 MHz, CDCl₃) δ –5.32 (br.).

phos‑peI‑25 (8). A 100 mL ﬂame‑dried Schlenck ﬂask

luted in 3 mL Et₂O, ﬁltered over Celite, and analyzed by GC.

equipped with a 50 mL dropping funnel was charged with a

solution of CDI, 0.485 g = 2.99 mmol, and NMM, 0.11 mL = Continuous Allylic Amination Catalysis

0.101 g = 1.0 mmol, ~1/3 equiv) in 5 mL of THF. A solution

A piece of Koch/SelRO MPF-50 membrane (stored in

of p-(diphenylphosphino)benzylamine (0.872 g = 2.99 mmol) EtOH) was cut to the correct size and pretreated by bathing

in THF (20 mL) was transferred into the dropping funnel and it in acetone overnight and subsequently in MeOH overnight.

subsequently slowly dropped into the ﬂask and allowed to re- The membrane was then installed in the reactor and ﬂushed

act for ~20 min. A solution of PEI-25 (0.187 g = 7.48 µmol, overnight with CH₂Cl₂and then with substrate solution (ap-

~1/400 equiv) in THF (30 mL) was then added in one portion prox. two reactor volumes). The substrate solution was pre-

Israel Journal of Chemistry

49

2009

95

pared by mixing 5.726 g crotyl methyl carbonate (0.044 mol),

4.182 g morpholine (0.048 mol) and 5.692 g n-decane (0.04

mol) in 400 mL CH₂Cl₂, transferred into the storage chamber

of the CFMR, and pumped through the reactor with a ﬂow rate

of 9 mL h^–1(150 µL min^–1). The catalyst solution was prepared

by dissolving [Pd(crotyl)Cl]₂(7.880 mg = 40 µmol of “Pd”)

and dendritic ligand (80 µmol of “P” if P/Pd = 2) in 4 mL

CH₂Cl₂and stirring for 3 h. The reaction was started by the

injection of 3 mL of catalyst solution into the membrane reac-

tor. Samples of 100 µL were taken at regular intervals (gener-

ally after one cycle = 40 min), quenched in 3 mL DBA/Et₂O,

ﬁltered over Celite, and analyzed by GC.

(h) Thunemann, A.F. Progr. Polym. Sci. 2002, 27(8),

1473–1572. (i) Kircheis, R.; Wightman, L.; Wagner, E.

Adv. Drug Delivery Rev. 2001, 53(3), 341–358. (j) Ber-

trand, P.; Jonas, A.; Laschewsky, A.; Legras, R. Macro-

mol. Rapid Comm. 2000, 21(7), 319–348. (k) Godbey,

W.T.; Wu, K.K.; Mikos, A.G. J. Controlled Rel. 1999,

60(2–3), 149–160.

(7) (a) Suh, J.; Cho, Y.; Lee, K.J. J. Am. Chem. Soc. 1991,

113(11), 4198–4202. (b) Lee, K.J.; Suh, J. Bioorg.

Chem. 1994, 22, 95–108. (c) Zoh, K.D.; Lee, S.H.; Suh,

J. Bioorg. Chem. 1994, 22, 242–252. (d) Kim, N.; Suh,

J. J. Org. Chem. 1994, 59(6), 1561–1571. (e) Suh, J.;

Lee, J.Y.; Hong, S.H. Bioorg. Med. Chem. Lett. 1997,

7(18), 2383–2386. (f) Suh, J.; Hah, S.S. J. Am. Chem.

Soc. 1998, 120(39), 10088–10093. (g) Suh, J.; Hong,

S.H. J. Am. Chem. Soc. 1998, 120(48), 12545–12552.

(h) Suh, J. Acc. Chem. Res. 2003, 36(7), 562–570.

(8) Shen, Z.; Chen, Y.; Frey, H.; Stiriba, S.-E. Macromol-

ecules 2006, 39(6), 2092–2099.

Continuous Allylic Thiolation Catalysis

A piece of Koch/SelRO MPF-50 membrane (stored in

EtOH) was cut to the correct size and rinsed in acetone over-

night and subsequently in MeOH overnight. The membrane

was then ﬁt in the reactor and ﬂushed overnight with CH₂Cl₂

and then with substrate solution (approx. two reactor vol-

umes). The substrate solution was prepared by mixing 2.290 g

crotyl methyl carbonate (0.0176 mol), 2.115 g thiophenol

(0.0192 mol), 0.324 g NMM (3.2 mmol) and 2.277 g n-dec-

ane (0.016 mol) in 160 mL CH₂Cl₂and was pumped through

the reactor with a ﬂow rate of 6 mL h^–1(100 µL min^–1). The

catalyst solution was prepared by dissolving [Pd(crotyl)Cl]₂

(7.880 mg = 40 µmol of “Pd”) and dendritic ligand (80 μmol

of “P” if P/Pd = 2) in 4 mL CH₂Cl₂and stirring for 3 h. The

reaction was started by the injection of 3 mL of catalyst so-

lution into the membrane reactor. Samples of 100 µL were

taken at regular intervals in time (generally after one cycle =

60 min), quenched in 3 mL DBA/Et₂O, ﬁltered over Celite,

and analyzed by GC.

(9) Krämer, M.; Pérignon, N.; Haag, R.; Marty, J.-D.;

Thomann, R.; Lauth-de Viguerie, N.; Mingotaud, C.

Macromolecules 2005, 38(20), 8308–8315.

(10) Hingst, M.; Tepper, M.; Stelzer, O. Eur. J. Inorg. Chem.

1998, 1, 73–82. For the use of the commercially avail-

able KPPh₂(THF sol., Aldrich) see: (a) Hay, A.M.;

Kerr, W.J.; Kirk, G.G.; Middlemiss, D. Organometal-

lics 1995, 14, 4986–4988. (b) Sablong, R.; Newton,

C.; Dierkes, P.; Osborn, J.A. Tetrahedron Lett. 1996,

37(28), 4933–4936. (c) Qiao, S.; Fu, G.C. J. Org.

Chem. 1998, 63, 4168–4169.

(11) Zhang, X.; Rodrigues, J.; Evans, L.; Hinkle, B.; Bal-

lantyne, L.; Peña, M. J. Org. Chem. 1997, 62(18),

6420–6423.

references And notes

(12) (a) Jansen, J.F.G.A.; Meijer, E.W.; de Brabander-van

den Berg, E.M.M. J. Am. Chem. Soc. 1995, 117(15),

4417–4418. (b) Stevelmans, S.; van Hest, J.C.M.; Jan-

sen, J.F.G.A.; van Boxtel, D.A.F.J.; de Berg, E.M.M.;

Meijer, E.W. J. Am. Chem. Soc. 1996, 118(31), 7398–

7399. (c) Baars, M.W.P.L.; Söntjens, S.H.M.; Fischer,

H.M.; Peerlings, H.W.I.; Meijer, E.W. Chem. Eur. J.

1998, 4(12), 2456–2466. (d) Schenning, A.P.H.J.; Elis-

sen-Román, C.; Weener, J.-W.; Baars, M.W.P.L.; van

der Gaast, S.J.; Meijer, E.W. J. Am. Chem. Soc. 1998,

120(32), 8199–8208. (e) Baars, M.W.P.L.; Kleppinger,

R.; Koch, M.H.J.; Yeu, S.-L.; Meijer, E.W. Angew.

Chem., Int. Ed. 2000, 39(7), 1285–1288. (f) Baars,

M.W.P.L.; Karlsson, A.J.; Sorokin, V.; de Waal, B.F.W.;

Meijer, E.W. Angew. Chem., Int. Ed. 2000, 39(23),

4262–4265. (g) Boas, U.; Karlsson, A.J.; de Waal,

B.F.M.; Meijer, E.W. J. Org. Chem. 2001, 66(6), 2136–

2145. (h) Pastor, L.; Barberá, J.; McKenna, M.; Marcos,

M.; Martín-Rapún, R.; Serrano, J.L.; Luckhurst, G.R.;

Mainal, A. Macromolecules 2004, 37(25), 9386–9394.

(i) McKenna, M.; Barberá, J.; Marcos, M.; Serrano, J.L.

J. Am. Chem. Soc. 2005, 127(2), 619–625.

(1) (a) Buhleier, E.; Wehner, W.; Vögtle, F. Synthesis 1978,

155. (b) Wörner, C.; Mülhaupt, R. Angew. Chem., Int.

Ed. 1993, 32(9), 1306–1308. (c) de Brabander-van den

Berg, E.M.M.; Meijer, E.W. Angew. Chem., Int. Ed.

1993, 32(9), 1308–1311.

(2) Reetz, M.T.; Lohmer, G.; Schwickardi, R. Angew.

Chem., Int. Ed. 1997, 36(13/14), 1526–1529.

(3) Brinkmann, N.; Giebel, D.; Lhmer, G.; Reetz, M.T.;

Kragl, U. J. Catal. 1999, 183, 163–168.

(4) Mizugaki, T.; Murata, M.; Ooe, M.; Ebitani, K.; Kane-

da, K. Chem. Commun. 2002, 52–53.

(5) Holter, D.; Frey, H. Acta Polym. 1997, 48, 298–309.

(6) (a) Krämer, M.; Stumbé, J.-F.; Grimm, G.; Krüger, U.;

Kaufmann, B.; Weber, M.; Haag, R. ChemBioChem

2004, 5, 1081–1087. (b) Aymonier, C.; Schlotterbeck,

U.; Antonietti, L.; Zacharias, P.; Thomann, R.; Tiller,

J.C.; Mecking, S. Chem. Commun. 2002, 3018–3019.

(c) Neu, M.; Fischer, D.; Kissel, T. J. Gene Med.

2005, 7(8), 992–1009. (d) Lungwitz, U.; Breunig, M.;

Blunk, T.; Goepferich, A. Eur. J. Pharm. Biopharm.

2005, 60(2), 247–266. (e) Piskin, E. Int. J. Pharm.

2004, 277(1–2), 105–118. (f) Kichler, A. J. Gene Med.

2004, 6(S1), S3–S10. (g) Pirrung, F.O.H.; Loen, E.M.;

Noordam, A. Macromol. Symp. 2002, 187, 683–693.

(13) (a) Girard, J.E.; Konaklieva, M.; Gu, J.; Guttman,

C.M.; Wetzel, S.J. Polym. Mater. Sci. Eng. Prepr. 2003,

88, 79. (b) Wang, D.-A.; Narang,A.S.; Kotb, M.; Gaber,

Ribaudo et al. / Functionalized Dendritic Phosphorus Ligands

96

A.O.; Miller, D.D.; Kim, S.W.; Mahato, R.I. Biomacro-

molecules 2002, 3(6), 1197–1207.

F.; van Leeuwen, P.W.N.M.; Reek, J.N.H. Top. Or-

ganomet. Chem. 2006, 20, 39–59. (y) Kassube, J.K.;

Gade, L.H. Top. Organomet. Chem. 2006, 20, 61–96.

(z) Hajij, C.; Haag, R. Top. Organomet. Chem. 2006,

20, 149–176.

(14) (a) de Groot, D.; Eggeling, E.B.; de Wilde, J.C.; Kooij-

man, H.; van Haaren, R.J.; van der Made, A.W.; Spek,

A.L.; Vogt, D.; Reek, J.N.H.; Kamer, P.C.J.; van Leeu-

wen, P.W.N.M. Chem. Commun. 1999, 1623–1624.

(b) de Groot, D.; Emmerink, P.G.; Coucke, C.; Reek,

J.N.H.; Kamer, P.C.J.; van Leeuwen, P.W.N.M. Inorg.

Chem. Commun. 2000, 3, 711–713. (c) de Groot, D.;

Reek, J.N.H.; Kamer, P.C.J.; van Leeuwen, P.W.N.M.

Eur. J. Org. Chem. 2002, 6, 1085–1095. (d) Reek,

J.N.H.; de Groot, D.; Oosterom, G.E.; van Heerbeek,

R.; Kamer, P.C.J.; van Leeuwen, P.W.N.M. Polym. Ma-

ter. Sci. Eng. 2001, 84, 158–159. (e) Oosterom, G.E.;

Steffens, S.; Reek, J.N.H.; Kamer, P.C.J.; van Leeuwen,

P.W.N.M. Top. Catal. 2002, 19, 61–73.

(16) (a) Tsuji, J.; Takahashi, J.; Morikawa, M. Tetrahedron

Lett. 1965, 6, 4387–4388. (b) Trost, B.M. Acc. Chem.

Res. 1980, 13, 385–393. (c) Tsuji, J. Tetrahedron 1986,

42, 4361–4401. (d) Trost, B.M.; van Vranken, D.L.

Chem. Rev. 1996, 96, 395–422. (e) Trost, B.M.; Craw-

ley, M.L. Chem. Rev. 2003, 103, 2921–2943. (f) Trost,

B.M. J. Org. Chem. 2004, 69, 5813–5837. (g) Nico-

laou, K.C.; Bulger, P.G.; Sarlah, D. Angew. Chem., Int.

Ed. 2005, 44, 4442–4489.

(17) (a) Johansen, M.; Jørgensen, K.A. Chem. Rev. 1998,

98, 1689–1708. (b) Hayashi, T.; Yamamaoto, A.; Ito,

Y.; Nishioka, E.; Miura, H.; Yanagi, K. J. Am. Chem.

Soc. 1989, 111(16), 6301–6311. (c) Trost, B.M. Angew.

Chem., Int. Ed. Eng. 1989, 28, 1173–1192. (d) Jumnah,

R.; Williams, J.M.J.; Williams, A.C. Tetrahedron Lett.

1993, 34, 6619. (e) Burgess, K.; Lium, L.T.; Pal, B. J.

Org. Chem. 1993, 58, 4758. (f) Trost, B.M.; van Vrank-

en, D.L. J. Am. Chem. Soc. 1993, 115, 444. (g) Magnus,

P.; Lacous, J.; Coldham, I.; Mugrage, B.; Bauta, W.B.

Tetrahedron 1995, 51, 11087. (h) Bower, J.F.; Jumnah,

R.; Williams, A.C.; Williams, J.M.J. J. Chem. Soc. Per-

kin Trans. 1 1997, 1411.

(18) (a) Feuerstein, M.; Laurenti, D.; Doucet, H.; Santelli,

M. Chem. Commun. 2001, 43–44. (b) Feuerstein, M.;

Laurenti, D.; Doucet, H.; Santelli, M. Tetrahedron Lett.

2001, 42, 2313–2315. (c) Feuerstein, M.; Laurenti, D.;

Doucet, H.; Santelli, M. J. Mol. Catal. A: Chem. 2002,

182–183, 471–480. (d) Mayer, H.A.; Kaska, W.C.

Chem. Rev. 1994, 94, 1239–1272. (e) Hierso, J.-C.;

Amardeil, R.; Bentabet, E.; Broussier, R.; Gautheron,

B.; Meunier, P.; Kalck, P. Coord. Chem. Rev. 2003, 236,

143–206.

(19) (a) Szabó, K.J. J. Am. Chem. Soc. 1996, 118, 7818–

7826. (b) Szabó, K.J. Organometallics 1996, 15, 1128–

1133. (c) Aranyos, A.; Szabó, K.J.; Castaño, A.M.;

Bäckvall, J.-E. Organometallics 1997, 16, 1058–1064.

(20) (a) Trost, B.M.; Toste, F.D. J. Am. Chem. Soc. 1999,

121(19), 4545–4554. (b) Delbecq, F.; Lapouge, C.

Organometallics 2000, 19, 2716–2723. (c) Branch-

adell, V.; Moreno-Mañas, M.; Pajuelo, F.; Pleixats, R.

Organometallics 1999, 18, 4934–4941. (d) Suzuki, T.;

Fujimoto, H. Inorg. Chem. 1999, 38(2), 370–382. (e)

Sakaki, S.; Nishikawa, M.; Ohyoshi, A. J. Am. Chem.

Soc. 1980, 102(12), 4062–4069. (f) Bigot, B.; Del-

becq, F. New J. Chem. 1990, 14, 659–669. (g) Szabó,

K.J.; Chem. Soc. Rev. 2001, 36, 136–143. (h) Hagelin,

H.; Åkermark, B.; Norrby, P.-O. Chem. Eur. J. 1999,

5(3) 902–909. (i) Robert, F.; Delbecq, F.; Nguefack,

C.; Sinou, D. Eur. J. Inorg. Chem. 2000, 351–358.

(j) Blöchl, P.E.; Togni, A. Organometallics 1996, 15,

4125–4132. (k) Hiroi, K.; Suzuki, Y.; Abe, I. Chem.

Lett. 1999, 149–150. (l) Chelucci, G.; Medici, S.; Saba,

A. Tetrahedron: Asymmetry 1999, 10, 543–550. (m)

(15) For reviews on dendritic catalysis see: (a) van Ko-

ten, G.; Jastrzebski, J.T.B.H. J. Mol. Catal. A: Chem.

1999, 146(1–2), 317–323. (b) Astruc, D.; Blaise, J.-C.;

Cloutet, E.; Diakovitch, L.; Rigaut, S.; Ruiz, J.; Sartor,

V.; Valerio, C. Top. Curr. Chem. 2000, 210, 229–259.

(c) Crooks, R.M.; Zhao, M.; Sun, L.; Chechik, V.;

Yeung, L.K. Acc. Chem. Res. 2001, 34(3), 181–190.

(d) Crooks, R.M.; Buford, I. III; Sun, L.; Yeung, L.K.;

Zhao, M. Top. Curr. Chem. 2001, 212, 81–135. (e)

Oosterom, G.E.; Reek, J.N.H.; Kamer, P.C.J.; van

Leeuwen, P.W.N.M. Angew. Chem., Int. Ed. 2001,

40(10), 1828–1849. (f) Astruc, D.; Chardac, F. Chem.

Rev. 2001, 101(9), 2991–3023. (g) Kreiter, R.; Kleij,

A.W.; Klein Gebbink, R.J.M.; van Koten, G. Top. Curr.

Chem. 2001, 217, 163–199. (h) Kriesel, J.W. Tilley,

T.D. Adv. Mater. 2001, 13(21), 1645-1648. (i) Twyman,

L.J.; King, A.S.H.; Martin, I.K. Chem. Soc. Rev. 2002,

31(2), 69–82. (j) Reek, J.N.H.; de Groot, D.; Oosterom,

G.E.; Kamer, P.C.J.; van Leeuwen, P.W.N.M. Rev. Mol.

Biotechnol. 2002, 90(3–4), 159–181. (k) Twyman, L.J.;

King, A.S.H. J. Chem. Res. 2002, (2), 43–59. (l) van

Heerbeek, R.; Kamer, P.C.J.; van Leeuwen, P.W.N.M.;

Reek, J.N.H. Chem. Rev. 2002, 102(10), 3717–3756.

(m) King, A.S.H.; Twyman, L.J. J. Chem. Soc. Perkin

Trans. 1, 2002, (20), 2209–2218. (n) Astruc, D. Pure

Appl. Chem. 2003, 75(4), 461–481. (o) Niu, Y.; Crooks,

R.M. C. R. Chimie, 2003, 6(8–10), 1049–1059. (p)

Reek, J.N.H.; de Groot, D.; Oosterom, G.E.; Kamer,

P.C.J.; van Leeuwen, P.W.N.M. C. R. Chimie, 2003,

6(8–10), 1061–1077. (q) Ribourdouille, Y.; Engel,

G.D.; Gade, L.H. C.R. Chimie, 2003, 6, 1087–1096.

(r) Chase, P.A.; Klein Gebbink, R.J.M.; van Koten, G.

J. Organomet. Chem. 2004, 689(24), 4016–4054. (s)

Scott, R.W.J.; Wilson, O.M.; Crooks, R.M. J. Phys.

Chem. B 2005, 109(2), 692–704. (t) Astruc, D.; Heuze,

K.; Gatard, S.; Mery, D.; Nlate S.; Plault, L. Adv. Synth.

Catal. 2005, 347(2–3), 329–338. (u) van de Coever-

ing, R.; Klein Gebbink, R.J.M.; van Koten, G. Progr.

Polym. Sci. 2005, 30(3–4), 474–490. (v) Caminade, A.-

M.; Mayoral, J.-P. Coord. Chem. Rev. 2005, 249(17–

18), 1917–1926. (w) Kofoed, J.; Reymond, J.-L. Curr.

Opin. Chem. Biol. 2005, 9(6), 656–664. (x) Ribaudo,

Israel Journal of Chemistry

49

2009

97

Albinati, A.; Pregosin, P.S.; Wick, K. Organometallics,

1996, 15(10), 2419–2421. (n) Steinhagen, H.; Reggelin,

M.; Helmchen, G. Angew. Chem., Int. Ed. Engl. 1997,

36(19), 2108–2110. (o) Junker, J.; Reif, B.; Steinhagen,

H.; Junker, B.; Felli, I.C.; Reggelin, M.; Griesinger, C.

Chem. Eur. J. 2000, 6(17), 3281–3286. (p) Dierkes,

P.; Ramdeehul, S.; Barloy, L.; De Cian, A.; Fischer, J.;

Kamer, P.C.J.; van Leeuwen, P.W.N.M.; Osborn, J.A.

Angew. Chem., Int. Ed. 1998, 37(22), 3116–3118.

(21) Vrieze, K.; van Leeuwen, P.W.N.M. Studies of Dynamic

Organometallic Compounds of the Transition Metals

by means of Nuclear Magnetic Resonance. In Progress

in Inorg. Chem. Vol. 14; Lippard, S.L., Ed.; Wiley and

Sons: New York, 1971.

(22) (a) Åkermark, B.; Zetterberg, K.; Hansson, S.; Kraken-

berger, B.; Vitagliano, A. J. Organomet. Chem. 1987,

335, 133–142. (b) Kurosawa, H. J. Organomet. Chem.

1987, 334, 243–253. (c) Åkermark, B.; Hansson, S.;

Vitagliano, A. J. Am. Chem. Soc. 1990, 112, 4587–

4588. (d) Frost, C.G.; Howarth, J.; Williams, J.M.J.

Tetrahedron: Asymmetry 1992, 3(9), 1089–1122. (e)

Kurosawa, H.; Ikeda, I. J. Organomet. Chem. 1992,

428, 289–301. (f) Pregosin, P.S.; Salzmann, R. Co-

ord. Chem. Rev. 1996, 155, 35–68. (g) Ramdeehul,

S.; Dierkes, P.; Aguado, R.; Kamer, P.C.J.; van Leeu-

wen, P.W.N.M.; Osborn, J.A. Angew. Chem., Int. Ed.

1998, 37(22), 3118–3121. (h) Kranenburg, M.; Kamer,

P.C.J.; van Leeuwen, P.W.N.M. Eur. J. Inorg. Chem.

1998, 25–27. (i) Hayashi, T.; Kawatsura, M.; Uozumi,

Y. J. Am. Chem. Soc. 1998, 120(8), 1681–1687. (j)

Gogoll, A.; Grennberg, H.; Axén, A. Organometallics

1998, 17(24), 5248–5253. (k) Amatore, C.; Gamez, S.;

Jutand, A. Chem. Eur. J. 2001, 7(6), 1273–1280. (l)

Fristrup, P.; Jensen, T.; Hoppe, J.; Norrby, P.-O. Chem.

653, American Chemical Society: Washington, DC,

1996. (c) Linford, L.; Raubenheimer, H.G.; Advances

in Organometallic Chemistry; Academic Press: San

Diego, 1991, Vol. 32, p 1. (d) Murray, S.G.; Hartley,

F.R. Chem. Rev. 1981, 81, 365. (e) Dubois, M.R. Chem.

Rev. 1989, 89, 1. (f) Kondo, T.; Mitsudo, T. Chem. Rev.

2000, 100, 3205–3220.

(28) Trost, B.M.; Scanlan, T.S. Tetrahedron Lett. 1986, 27,

4141.

(29) (a) Goux, C.; Lhoste, P.; Sinou, D. Tetrahedron Lett.

1992, 33, 8099. (b) Goux, C.; Lhoste, P.; Sinou, D. Tet-

rahedron 1994, 50, 10321.

(30) (a) Koch Membrane Systems, Wilmington, DE, USA,

http:// www.kochmembrane.com; (b) Linder, C.; Ne-

mas, M.; Perry, M.; Katraro, R. EP 532199, 1993.

(31) Eggeling, E.B.; Hovestad, N.J.; Jastrzebski, J.T.B.H.;

Vogt, D.; van Koten, G. J. Org. Chem. 2000, 65, 8857–

8865.

(32) Kleij, A.W.; Gossage, R.A.; Klein Gebbink, R.J.M.;

Brinkmann, N.; Reijerse, E.J.; Kragl, U.; Lutz, M.;

Spek, A.L.; van Koten, G. J. Am. Chem. Soc. 2000,

122, 12112–12124.

(33) Dijkstra, H.P.; Kruithof, C.A.; Ronde, N.; van de Co-

evering, R.; Ramón, D.J.; Vogt, D.; van Klink, G.P.M.;

van Koten, G. J. Org. Chem. 2003, 68, 675.

(34) (a) Dick, C.R.; Ham, G.E. J. Macromol. Sci. Chem.

A 1970, 4, 1301. (b) van den Berg, J.W.A.; Bloys van

Treslong, C.J.; Polderman, A. Rec. Trav. Chim. 1973,

92, 3. (c) Lukovkin, G.M.; Pshezhetsky, V.S.; Mur-

tazaeva, G.A. Eur. Polym. J. 1973, 9, 559–565. (d)

Park, I.H.; Choi, E.-J. Polymer 1996, 37(2), 313–319.

(e) Pfau, A.; Schrepp, W.; Horn, D. Langmuir 1999,

15(9), 3219–3225.

(35) (a) Gopidas, K.R.; Leheny, A.R.; Caminati, G.; Turro,

N.J.; Tomaila, D.A. J. Am. Chem. Soc. 1991, 113(19),

7335–7342. (b) Young, J.K.; Baker, G.R.; Newkome,

G.R.; Morris, K.F.; Johnson, C.S.Jr. Macromolecules

1994, 27(13), 3464–3471. (c) Hawker, C.J.; Malm-

ström, E.E.; Frank, C.W.; Kampf, J.P. J. Am. Chem.

Soc. 1997, 119(41), 9903–9904. (d) Harth, E.M.;

Hecht, S.; Helms, B.; Malmström, E.E.; Fréchet,

J.M.J.; Hawker, C.J. J. Am. Chem. Soc. 2002, 124(15),

3926–3938. (e) Passeno, L.M.; Mackay, M.E.; Baker,

G.R.; Vestberg, R.; Hawker, C.J. Macromolecules

2006, 39(2), 740–746. (f) Christensen, J.B.; Nielsen,

M.F.; van Haare, J.A.E.H.; Baars, M.W.P.L.; Janssen,

R.A.J.; Meijer, E.W. Eur. J. Org. Chem. 2001,(11),

2123–2128. (g) Cheruvalill, S.R.; Capitosti, C.J.;

Cramer, S.J.; Modarelli, D.A. J. Phys. Chem. B 2001,

105(42), 10175–10188. (h) Gronheid, R.; Stefan, A.;

Cotlet, M.; Hofkens, J.; Qu, J.; Müllen, K.; van der

Auweraer, M.; Verhoeven, J.W.; de Schryver, F.C.

Angew. Chem., Int. Ed. 2003, 42(35), 4209–4214.

(i) Chasse, T.L.; Sachdeva, R.; Li, Q.; Li, Z.; Petrie,

R.J.; Gorman, C.B. J. Am. Chem. Soc. 2003, 125(27),

8250–8254. (j) Thomas, K.R.J.; ThompsonA.L.; Siva-

kumar, A.V.; Bardeen, C.J.; Thayumanavan, S. J. Am.

Eur. J. 2006, 12, 5352–5360

.

(23) Ribourdouille, Y.; Engel, G.D.; Richard-Plouet, M.;

Gade, L.H. Chem. Commun. 2003, 1228–1229.

(24) (a) Hayashi, T. Catalytic Asymmetric Synthesis, Ed.

I. Ojima, VCH, Weinheim, 1993, p 325. (b) Helm-

chen, G.; Pfaltz, A. Acc. Chem. Res. 2000, 33, 336. (c)

Heumann, A. Transition Metals for Organic Synthesis;

Beller, M.; Bolm, C., Eds. Wiley-VCH: Weinheim,

1998, p 251.

(25) (a) Zwanenburg, B.; Klunder, A.J.H. Perspectives in

Organic Chemistry of Sulfur; Elsevier: Amsterdam,

1987. (b) Thuillier, A.; Metzner, P. Sulfur Reagents

in Organic Synthesis; Academic Press: New York,

1994.

(26) (a) Hegedus, L.L.; McCabe, R.W. Catalyst Poison-

ing; Marcel Dekker: New York, 1984. (b) Hutton, A.T.

Comprehensive Coordination Chemistry; Wilkinson,

G., Gillard, A.D., McCleverty, J.A., Eds.; Pergamon:

Oxford, 1984, Vol. 5, p 1151.

(27) (a) Transition Metal Sulphides; Weber, T., Prins, R.,

van Santen, R.A., Eds.; Kluwer Academic Publishers:

Dordrecht, 1998. (b) Transition Metal Sulfur Chem-

istry: Biological and Industrial Signiﬁcance; Stiefel,

E.I., Matsumoto, K., Eds.; ACS Symposium Series

Chem. Soc. 2005, 127(1), 373–383

.

Ribaudo et al. / Functionalized Dendritic Phosphorus Ligands

98

(36) (a) Naylor, A.M.; Goddard, W.A.,III. J. Am. Chem. Soc.

(39) (a) Scherrenberg, R.; Coussens, B.; van Vliet, P.; Ed-

ouard, G.; Brackman, J.; de Brabander, E.; Mortensen,

K. Macromolecules 1998, 31(2), 456–461. (b) Bodnár,

I.; Silva, A.S.; Deitcher, R.W.; Weisman, N.E.; Kim,

Y.G.; Wagner, N.J. J. Polym. Sci. B: Polym. Phys. 2000,

38(6), 857–873. (c) Rosenfeldt, S.; Dingenouts, N.;

Ballauff, M.; Werner, N.; Vögtle, F.; Lindner, P. Macro-

molecules 2002, 35(21), 8098–8105. (d) Stancik, C.M.;

Pople, J.A.; Trollsås, M.; Lindner, P.; Hedrick, J.L.;

Gast, A.P. Macromolecules 2003, 36(15), 5765–5775.

(e) Jana, C.; Jayamurugan, G.; Ganapathy, R.; Maiti,

P.K.; Jayaraman, N.; Sood, A.K. J. Chem. Phys. 2006,

124(20), 204719 (1–10).

(40) (a) Burchard, W. Adv. Polym. Sci. 1999, 143, 113–191.

(b) Sheiko, S.S.; Moller, M. Top. Curr. Chem. 2001,

212, 137–175. (c) Seiler, M. Fluid Phase Equilib. 2006,

241(1–2), 155–174. (d) Hanselmann, R.; Hölter, D.;

Frey, H. Macromolecules 1998, 31(12), 3790–3801.

(e) von Ferber, Ch.; Jusuﬁ, A.; Watzlawek, M.; Likos,

C.N.; Löwen, H. Phys. Rev. E 2000, 62(5), 6949–6956.

(f) Likos, C.N.; Schmidt, M.; Löwen, H.; Ballauff,

M.; Pötschke, D.; Lindner, P. Macromolecules 2001,

34(9), 2914–2920. (g) Likos, C.N.; Rosenfeldt, S.;

Dingenouts, N.; Ballauff, M.; Lindner, P.; Werner, N.;

Vögtle, F. J. Chem. Phys. 2002, 117(4), 1869–1877. (h)

Jurjiu, A.; Koslowski, Th.; von Ferber, Ch.; Blumen, A.

Chem. Phys. 2003, 294(2), 187–199. (i) Liu, H.; Wilén,

C.-E. J. Polym. Sci. B: Polym. Phys. 2004, 42(7), 1235–

1242. (j) Zhou, Z.; Yu, M.; Yan, D.; Li, Z. Macromol.

Theory Simul. 2004, 13(8), 724–730. (k) Burchard, W.

Macromolecules 2004, 37(10), 3841–3849. (l) Buzza,

D.M.A. Eur. Phys. J. E 2004, 13(1), 79–86. (m) Ko-

slowski, Th.; Jurjiu, A.; Blumen, A. Macromol. Theory

Simul. 2006, 15(7), 538–545.

1989, 111(6), 2339–2341. (b) Lescanec, R.L.; Muthu-

kumar, M. Macromolecules 1990, 23(8), 2280–2288.

(c) Miklis, P.; Çağin, T.; Goddard, W.A.,III, J. Am.

Chem. Soc. 1997, 119(32), 7458–7462. (d) Cavallo,

L.; Fraternali, F. Chem. Eur. J. 1998, 4(5), 927–934.

(e) Rosini, C.; Superchi, S.; Peerlings, H.W.I.; Meijer,

E.W. Eur. J. Org. Chem. 2000, (1), 61–71. (f) Zhang,

X.; Haxton, K.J.; Ropartz, L.; Cole-Hamilton, D.J.;

Morris, R.E. J. Chem. Soc. Dalton Trans. 2001, (22),

3261–3268. (g) Mackay, M.E.; Carmezini, G. Chem.

Mater. 2002, 14(2), 819–825. (h) Zacharopoulos, N.;

Economou, I.G. Macromolecules 2002, 35(5), 1814–

1821. (i) Versluis, C.; Hillegers, T. Macromol. Theory

Simul. 2002, 11(6), 640–648. (j) Götze, I.O.; Likos,

C.N. Macromolecules 2003, 36(21), 8189–8197. (k)

Jasch, F.; von Ferber, Ch.; Blumen, A. Phys. Rev. E,

2003, 68(5), 051106 (1–12). (l) Blumen, A.; von Ferber,

Ch.; Jurjiu, A.; Koslowski, Th. Macromolecules 2004,

37(2), 638–650. (m) Ballauff, M.; Likos, C.N. Angew.

Chem., Int. Ed. 2004, 43(23), 2998–3020. (n) Blumen,

A.; Jurjiu, A.; Koslowski, Th.; Friedrich, Ch. Macro-

mol. Symp. 2006, 237(1), 53–59. (o) Satmarel, C.; von

Ferber, Ch.; Blumen, A. J. Chem. Phys. 2006, 124(17),

174905 (1–10).

(37) (a) Tande, B.M.; Wagner, N.J.; Mackay, M.E.; Hawker,

C.J.; Jeong, M. Macromolecules 2001, 34(24), 8580–

8585. (b) De Luca, E.; Richards, R.W. J. Polym. Sci. B:

Polym. Phys. 2003, 41(12), 1339–1351. (c) Radke, W.;

Müller, A.H.E. Macromolecules 2005, 38(9), 3949–

3960.

(38) (a) Rosenfeldt, S.; Dingenouts, N.; Ballauff, M.; Lind-

ner, P.; Likos, C.N.; Werner, N.; Vögtle, F. Macromol.

Chem. Phys. 2002, 203(13), 1995–2004. (b) Kunam-

anemi, S.; Buzza, D.M.A.; Parker, D.; Feast, W.J. J.

Mater. Chem. 2003, 13(11), 2749–2755.

(41) Tsuji, J.; Sato, K.; Okumoto, H. J. Org. Chem. 1984,

49(8), 1341–1344.

Israel Journal of Chemistry

49

2009

Article Doi

Phosphorus functionalized dendrimers and hyperbranched polymers: Is there a need for perfect dendrimers in catalysis?

DOI: 10.1560/IJC.49.1.79

Source and publish data:

Authors:

Article abstract of DOI:10.1560/IJC.49.1.79

Full text of DOI:10.1560/IJC.49.1.79

Products guided by the article

R&D Labs maybe for 1203457-76-4

Relevant to this article

Hot Product