New cyclic phosphonium salts derived from the reaction of phosphine-aldehydes with acid

Article abstract of DOI:10.1016/j.jorganchem.2010.04.016

Various cyclic phosphonium structures are formed in high yield by the deprotection of unstable phosphine-aldehydes in acidic solution. When there is a methylene spacer between the phosphine and the aldehyde, a phosphonium ion [PHR₂CH₂CH(OEt)₂]Br₂, R=iPrOH, Et is obtained. Reaction of these phosphonium salts with water produces the dimers [-PR₂CH₂CH(OH)-]₂[Br]₂?R = iPr, Et. When there is an ethylene spacer as in PPh₂CH₂CH₂CH(OCH₂CH₂O), a remarkable tetramer with a 16-membered ring [-PPh₂CH₂CH₂CH (OH)-]₄[Cl]₄ forms as one diastereomer in hydrochloric acid solution. Reaction of HCl with the protected phosphine-aldehyde with a propylene spacer (PPh₂CH₂CH₂CH₂CH(OCH₂CH₂O)) results in the formation of the monomeric phosphonium salt [-PPh₂ CH₂CH₂CH₂CH(OH)-]Cl with a 5-membered ring. Solid state structures of different ring types were determined using X-ray diffraction experiment.

Full text of DOI:10.1016/j.jorganchem.2010.04.016

Journal of Organometallic Chemistry 695 (2010) 1824e1830
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
New cyclic phosphonium salts derived from the reaction of phosphine-aldehydes
with acid
*
Alexandre A. Mikhailine, Paraskevi O. Lagaditis, Peter E. Sues, Alan J. Lough, Robert H. Morris
Department of Chemistry, University of Toronto, 80 St. George St., Toronto, ON, M5S 3H6, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
Various cyclic phosphonium structures are formed in high yield by the deprotection of unstable phos-
phine-aldehydes in acidic solution. When there is a methylene spacer between the phosphine and the
aldehyde, a phosphonium ion [PHR2CH2CH(OEt)2]Br2, R=iPrOH, Et is obtained. Reaction of these phos-
phonium salts with water produces the dimers [ePR2CH2CH(OH)e]2[Br]2 R ¼ iPr, Et. When there is an
ethylene spacer as in PPh2CH2CH2CH(OCH2CH2O), a remarkable tetramer with a 16-membered ring
[ePPh2CH2CH2CH (OH)e]4[Cl]4 forms as one diastereomer in hydrochloric acid solution. Reaction of HCl
with the protected phosphine-aldehyde with a propylene spacer (PPh2CH2CH2CH2CH(OCH2CH2O))
results in the formation of the monomeric phosphonium salt [ePPh2 CH2CH2CH2CH(OH)e]Cl with a 5-
membered ring. Solid state structures of different ring types were determined using X-ray diffraction
experiment.
Received 30 March 2010
Received in revised form
10 April 2010
Accepted 17 April 2010
Available online 24 April 2010
Keywords:
Ligand design
Phosphorus
Macrocycles
Aldehyde
Ó 2010 Elsevier B.V. All rights reserved.
Deprotection
1. Introduction
less stable phosphine-aldehydes derived by reactions of base with
dimeric phosphonium salts 1 can also be used for the synthesis of
Phosphines are particularly versatile ligands in organometallic
chemistry because of the ease in changing the substituents at
phosphorus to vary their electronic and steric characteristics [1].
Additional donor groups can be attached to make mixed donor
chelating ligands such as phosphorusenitrogen [2] or phosphor-
useoxygen ligands [3]. On the other hand, the synthesis and
reactions of molecules with a phosphine group and an electrophilic
group such as an aldehyde are less explored. Such amphoteric
molecules may undergo unselective polymerization by the forma-
tion of carbonephosphorus bonds and for this reason they require
specific handling and are difficult to prepare in good yield and high
purity [4]. Matt and co-workers [4a] observed that under acidic
conditions a particular amphoteric phosphine-aldehyde selectively
forms a stable phosphonium dimer (1, Scheme 1, M ¼ Li).
iron(II) catalysts [7]. These precatalysts were found to be particularly
active in the enantioselective transfer hydrogenation of ketones
using isopropanol as the reducing agent [8].
The ease of formation of the phosphorusecarbon bonds in the
synthesis of dication 1 might also be exploited in the synthesis of
macrocyclic molecules (Scheme 1) to produce porous materials,
anionic sensors or macrocycles. In recent years macrocyclic mole-
cules have attracted more attention from the chemical community
as a result of developments in the areas of molecular signaling,
molecular motors and organometallic catalysis [9]. Macrocycles are
commonly synthesized from linear molecules containing terminal
reactive functional groups or from short amphoteric molecules that
oligomerize into cyclic structures. Both approaches demand a high
level of selectivity towards the formation of cyclic molecules over
the linear ones and towards a specific size of the ring. One of the
solutions to the selectivity problem is a synthesis involving
a template pre-organization [10]. Here a templating molecule
brings together two or more molecules with reactive functionalities
by noncovalent interactions to form a highly organized structure
that further undergoes chemoselective reactions.
A particularly useful phosphine-aldehyde, ortho-diphenylphos-
phinobenzaldehyde, is the starting material for the synthesis of chiral
PNNP and PNN ligands by Schiff base condensation reactions with
enantiopure amines and diamines [5]. Transition metal complexes of
these ligands have found to be highly efficient and enantioselective
catalysts [6]. The reason for the popularity of this phosphine-alde-
hyde relative to others is its stability. Recently our group showed that
Templating molecules are typically cationic because their coor-
dination chemistry is well studied and understood [11]. Anions are
less often employed in template syntheses because of their unde-
sirable features, which include a high solvation energy, a low
charge to radius ratio and a high pH dependence. Several examples
* Corresponding author. Tel.: þ1 416 978 6962.
E-mail address: rmorris@chem.utoronto.ca (R.H. Morris).
0022-328X/$ e see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2010.04.016

A.A. Mikhailine et al. / Journal of Organometallic Chemistry 695 (2010) 1824e1830
1825
to give an intermediate phosphonium salt that was then hydro-
lyzed (Scheme 2).
The yields of the white solids 2a and 2b are 81 and 40%,
respectively. Compound 2a is an air- and moisture-stable white
solid that is completely soluble in methanol and/or water and
insoluble in other common organic solvents. Therefore it possesses
similar physical properties compared to the dimer 1. The dimer 2b,
on the other hand, adsorbs water on prolonged exposure to the air,
and is soluble only in water. However an aqueous solution of 2b is
stable toward oxidation by molecular oxygen.
This direct reaction does not work for the less nucleophilic
phosphines such as diphenylphosphine (HPPh2) and di(p-tolyl)
phosphine (HP(p-tol)2). These need to be converted into their cor-
responding more nucleophilic phosphides by reaction with potas-
sium hydride before a successful nucleophilic attack on the
protected aldehyde BrCH2CH(OEt)2 can take place (see Schemes 1
and 3). Deprotection of the phosphine-aldehyde diethyl acetals in
acidic media gives the desired dimers with diphenyl (1, Scheme 1)
or di-p-tolyl substituents (2c, Scheme 3). The new dimeric
compound 2c was obtained as a white, air-stable solid in 88% yield.
Hence the high yields and chemoselectivity in the formation of 1
and 2 are not affected by varying the electronic and steric proper-
ties of the phosphines. An explanation for the selectivity must lie in
the favorable geometry of the 6-membered rings. Thus, the selec-
tivity of the reaction appears to be primarily controlled by ther-
modynamic factors rather than kinetic ones.
Scheme 1. Synthesis and behavior of phosphine-aldehydes under different pH
conditions.
The new dimers 2a and 2c were fully characterized by NMR
spectroscopy and X-ray diffraction experiments. These compounds
show two characteristic singlets in the 31P{1H} NMR spectra in the
region between 11 and 40 ppm; the two singlets observed in the 31P
{1H} NMR arise from the rac and meso diastereomers. As for the 1H
NMR spectra there is a characteristic multiplet in the region between
5.3 and 6.2 ppm for the proton on the carbon with the hydroxyl
group eCH(OH)e, and an absence of downfield shift of the aldehyde
hydrogen resonance which is expected in the 9e10 ppm region.
Crystals of 2a and 2c have been analyzed by single crystal X-ray
diffraction (Fig. 1). Both structures are meso diastereomers with the
hydroxyl groups in equatorial positions for 2a and in axial positions
for 2c. The PeC1 bond lengths are similar in both dimers and are
slightly elongated compared to all other phosphorusecarbon bonds
in the structures probably as a result of electron-withdrawing effect
of the neighbouring oxygen atom. The bond between P1 and C6 in
2a is longer than the bond between P1 and C3 in 2c as expected
from the larger cone angle of the isopropyl group compared to
p-tolyl group. All of the angles and bond lengths of both 2a and 2c
are comparable to those observed by Matt and co-workers [4a].
Crystals that were used for the X-ray diffraction studies were
also analyzed by 31P NMR spectroscopy in order to assign peaks and
determine the ratio of diastereomers formed in the reaction. The
31P {1H} NMR spectra obtained for the crystals showed peaks for
both diastereomers at 36.8 and 34.5 ppm for 2a and at 16.1 and
11.2 ppm for 2c with integration ratios of 1:2 and 3:4, respectively.
On the other hand Matt and co-workers were able to selectively
of anion-assisted formation of rotaxanes and similar macromole-
cules are found in the literature [12]. The only multi-component
anionic template to our knowledge was reported by Rubio and co-
workers [12] in the synthesis of macrocycles derived from amino
acids. Unfortunately this methodology usually gives low to
moderate yields of the desired product.
The ease of formation of the dimeric phosphonium compound,
1, (Scheme 1) instead of other cyclic oligomers prompted us to
explore this reaction in more detail. We wanted to investigate: (a)
how is the chemoselectivity and diastereoselectivity controlled in
the process of the formation of dimer 1 and (b) whether this
knowledge could be used to prepare more complex and diverse
molecules. This led us to discover an even simpler method of
producing such phosphonium compounds when the phosphorus
atoms bear electron-donating substituents and a high yielding
method for the selective formation of a tetrameric macrocycle.
2. Results and discussion
2.1. One carbon spacer between the phosphorus and reactive
carbon centre
Cyclic phosphonium dimers similar to that of 1 but with iso-
propyl or ethyl substituents at the phosphorus atoms (2a, 2b) were
prepared in a new, direct reaction of the secondary phosphine with
the protected bromoacetoaldehyde diethyl acetal (BrCH2CH(OEt)2)
Scheme 2. Direct preparation of phosphonium dimers 2a and 2b using secondary phosphines containing electron-donating alkyl substituents.

1826
A.A. Mikhailine et al. / Journal of Organometallic Chemistry 695 (2010) 1824e1830
Scheme 3. General preparation of phosphonium dimers 1 and 2c from potassium phosphide.
crystallize dimer 1 as the meso diastereomer and determine the
structure by X-ray diffraction. A signal at 16.6 ppm in the 31P {1H}
NMR spectrum was observed thus indicating that it is the minor
diastereomer in the mixture of 1. A possible explanation for the
difference in distribution of diastereomers of 2a and 2c resulting
from X-ray diffraction and 31P {1H} NMR analyses is that the dia-
stereomers have similar solubility and both diastereomers co-
crystallize. Thus the crystals containing the meso diastereomer
were chosen for the X-ray diffraction accidentally. It is however
possible that racemization of the pure meso diastereomer occurred
in methanol-d4 into a mixture of meso and rac diastereomers prior
to 31P {1H} NMR analysis.
In order to test each of those assumptions, equimolar amounts of
1 and 2a were mixed together in methanol-d4 and the solution was
monitored by 31P {1H} NMR spectroscopy. After 12 h a spectrum
showed only peaks at 16.63 and 11.97 ppm (dimer 1) and at 36.80
and 34.54 ppm (dimer 2a). Since no mixed-substituent phospho-
nium dimer was produced, this indicated that the dissociation of
dimers or racemization does not take place in solution (Table 1).
The various substituents at the phosphorus atom have a minor
effect on the chemoselectivity of the reaction and six-membered
dimers are formed over other possible species (Scheme 1).
Conversely, more electron-rich substituents (isopropyl and ethyl)
considerably increase the formation of one diastereomer over the
other and thus make the reaction more diastereoselective. One can
argue that the selectivity comes from the formation of the more
thermodynamically favorable six-membered ring instead of the
less stable cyclic or linear oligomer and polymer. In order to
investigate such an assumption, we decided to increase the size of
the hydrocarbon spacer between the carbonyl and phosphine
groups in the phosphine-aldehyde and study the oligomerization
under acidic conditions.
2.2. Two carbon spacer between the phosphorus and reactive
carbon centre
The synthesis of 2-(2-chloroethyl)-1,3-dioxolane was previously
reported by the Shrestha-Dawadi and Lugtenburg [13]. The addi-
tion of the b-chloropropionaldehyde ethylene glycol to a freshly
prepared solution of the KPPh2 in THF resulted in the formation of
the tertiary phosphine 3 with a protected aldehyde in situ. Instead,
the addition of a degassed, concentrated, aqueous hydrochloric acid
Table 1
Selected spectroscopic data and properties of the dimers 1, 2a, 2b, 2c.
Dimer 31P{1H}
NMR minor
31P{1H}
NMR major
Ratio of
diastereomers total
Yield (%)
diastereomer (ppm) diastereomer (ppm) minor/majora
Fig. 1. ORTEP views of the molecular structures of 2a (top) and 2c (bottom). Thermal
ellipsoids are drawn at 30% probability. Some hydrogen atoms, solvent molecules and
counter ions are omitted for clarity. Selected interatomic distances [Å] and angles [o]:
2a P1C1 1.835(3), P1C2 1.813(3), P1C6 1.825(3), P1C3 1.819(3), C1O1 1.410(3); C1P1C2
103.7(1), C1P1C3 108.4 (1); 2c P1C1 1.847(5), P1C2 1.804(4), P1C3 1.783(3), C1O1 1.394
(4); C1P1C2 104.7(2), C10P1C3 109.1(2).
1[4a]
2a
2b
16.6
36.8
35.5
16.1
11.9
34.5
32.7
11.2
3:4
1:2
1:2
3:4
93
81
87
88
2c
a
Ratios of diastereomers were determined by 1H NMR.

A.A. Mikhailine et al. / Journal of Organometallic Chemistry 695 (2010) 1824e1830
1827
Commercially
available
2-(3-chloropropyl)-1,3-dioxolane
(ClCH2CH2CH2CH(OCH2CH2O)) was reacted with a freshly prepared
solution of KPPh2 in THF. Quenching of the reaction with concen-
trated hydrochloric acid gave rise to a species in solution that
produced a strongly downfield-shifted resonance in the 31P{1H}
NMR spectrum at 38.0 ppm. An X-ray diffraction study revealed
that 6 is a monomeric phosphonium salt. Here the molecule has
clearly “bitten its tail” to form a stable five-membered ring struc-
ture (Fig. 2). The structure of 6 revealed that bond lengths in the
molecule are in comparable range to those observed in the previous
cyclic oligomers. However, the C1eP1eC4 angle is smaller in
compound 6 which is consistent with the observed five-membered
cyclic geometry.
3. Conclusion
In conclusion, a series of oxygen-stable, cyclic phosphonium
salts, derived from unstable phosphine aldehydes were synthesized
and fully characterized. Chemoselective cyclization reactions
strongly depended on the stability of the final product, although in
the case of tetramer 4, an anionic template effect might be opera-
tional. The variation of the carbon distance between phosphorus
and carbonyl carbon from 2 to 4 resulted in the formation of
different phosphonium salts: dimers with six-membered rings
(compounds 1 and 2), a tetramer with a 16-membered ring
Scheme 4. Synthesis of tetramer 4 from 2-(2-chloroethyl)-1,3-dioxolane and potas-
sium phosphide.
solution to the reaction mixture gave 4 as a white precipitate
(Scheme 4).
(compound 4) and
a monomer with a five-membered ring
(compound 6), respectively. All of the phosphine-aldehydes react in
the presence of acid to form carbon-phosphorus bonds chemo-
selectively. High diastereoselectivity of oligomerization was
observed in the process of formation of the dimers with electron-
donating alkyl substituents at phosphorus (dimers 2a and 2b) and
with the tetrameric oligomer 4. The high chemoselectivity that was
observed is controlled by the thermodynamic stability of the
structures formed compared to other possible species (Scheme 1).
This conclusion arises from observations that the dimers do not
dissociate in solution into monomers and that their yields are not
effected by changes in the reaction conditions as would be expected
in case of kinetic control.
The molecular structure of the isolated white solid was deter-
mined by a preliminary X-ray diffraction experiment (see Experi-
mental section) to be cyclic and tetrameric (Scheme 4). The
elemental analysis and 1H NMR spectra show that the crystals
contain approximately 3 water molecules per tetramer. All attempts
to completely remove water from the solid tetramer failed, probably
due to the high affinity of the chloride ions for water. Since the 31P
{1H} NMR spectrum of 4 in methanol-d4 solution shows a single
resonance at 27.08 ppm, we may conclude the reaction leading to its
tetrameric structure is highly diastereoselective.
The formation of the specific ring size of 4 might suggest the
selectivity toward its formation may be due to a template effect of
monomers around the chloride ion. If this assumption is correct
then the use of a more bulky counter ion in the synthesis should
lead to the formation of a larger ring. To check this assumption, we
synthesized the phosphine-aldehyde monomer PPh2CH2CH2CH2C
(O)H (5) by reacting the tetramer with base and extracting the
product with diethyl ether. Spectral data of compound 5 were
found to be similar to the one reported by Vaughn and Gladysz [4d]
who synthesized and characterized this compound previously by
a different method.
The reaction of 5 in THF with hydrochloric acid resulted in the
complete conversion of the monomer to the tetramer 4. Similar
reactions were performed with HBr, HBF4 and p-TsOH in methanol-
d4. The 31P{1H} NMR spectra of the reaction mixtures showed
complete consumption of the starting phosphine 5 and production
of the tetramer 4 along with other oligomeric species as minor
products. Attempts to vary the conditions of the reaction in order to
increase the yield of non-tetrameric species such as by changing
the solvent, temperature and rate of addition of the reagents, failed
probably as a result of the high stability of the tetramer 4 relative to
the other oligomers.
4. Experimental
4.1. General considerations
All procedures and manipulations involving air-sensitive
materials were performed under an argon or nitrogen atmosphere
using Schlenk techniques or a glovebox with N2 (g). Solvents were
degassed and dried using standard procedures prior to all manip-
ulations and reactions. All other reagents used in the procedures
were purchased from commercial sources and utilized without
further purifications. NMR spectra were recorded at ambient
temperature and pressure using Varian Gemini 400 MHz and
300 MHz spectrometers [1H (400 MHz and 300 MHz), 13C{1H}
(100 MHz and 75 MHz) and 31P {1H} (161 MHz and 121 MHz)]. 1H
and 13C{1H} were referenced to solvent resonances. Elemental
analyses were done at the University of Toronto on a PerkineElmer
2400 CHN elemental analyzer.
4.2. General procedure for preparation of the dimers 2a and 2b
In an Ar-filled glovebox diisopropylphosphine (for 2a) or
diethylphosphine (for 2b) (15 mmol) was dissolved in 5 mL of dry
THF. Bromoacetadehyde diethyl acetal (15 mmol) was added to the
resulting mixture on an Ar Schlenk line and the resulting solution
was stirred for 4 h. The reaction was quenched with degassed H2O
(20 mmol) and heated for overnight at 70 ꢀC. The solvent was
2.3. Three carbon spacer between the phosphorus and reactive
carbon centre
The synthesis of the protected phosphine-aldehyde precursor was
carriedoutinasimilarfashiontothepreviouscompounds (Scheme 5).

1828
A.A. Mikhailine et al. / Journal of Organometallic Chemistry 695 (2010) 1824e1830
Scheme 5. Synthesis of compound 6 from 2-(3-chloropropyl)-1,3-dioxolane and potassium phosphide.
partially removed under vacuum and to give a colorless solution
with a white precipitate. The solution was stored for 3 h at 5 ꢀC. The
precipitate then was filtered and washed with diethyl ether
(2 ꢁ 5 mL) to give an analytically pure sample. Crystals suitable for
X-ray diffraction experiments were obtained by slow diffusion of
diethyl ether into a saturated solution of 2a in methanol.
diastereomers). Anal. Calcd for C16H36P2O2Br2: C, 39.85; H, 7.52.
Found: C, 39.35; H, 7.32.
4.2.2. Dimer 2b
Yield: 2.78 g, 87%. The diastereomeric ratio was found to be 1:2,
as determined by 1H NMR. 1H NMR (400 MHz, D2O with a CDCl3/
TMS insert, resonances of two diastereomers of 2b overlap; see
4.2.1. Dimer 2a
below):
{
d
5.45e5.10 (m, 2H, CH(OH), diastereomers overlap; 1H
2
Yield: 2.93 g, 81%. The diastereomeric ratio was found to be
31P}, 5.36 (pseudo d, 2H, CH(OH), JH-P ¼ 5.8 Hz, major diaste-
1:2, as determined by 1H NMR. 1H NMR (400 MHz, CD3OD, reso-
reomer), 5.29 (pseudo dd, 2H, CH(OH), 3JH-H ¼ 3.4 Hz, 2JH-P ¼ 9.2 Hz,
minor diastereomer)), 3.31e2.87 (m, 4H, CH(OH)CH2P, overlap of
diastereomers; 1H{31P}, same), 2.58e2.08 (m, 8H, (CH3CH2P, over-
lap of diastereomers; 1H{31P}; same), 1.32e0.90 (m, 12H, CH3CH2P,
overlap of diastereomers; 1H{31P}, same). 31P{1H} NMR (161 MHz,
nances of two diastereomers of 2a overlap in region
d 3.50e1.40;
3
2
see below):
d
5.60 (pseudo ddd, JHH ¼ 6.4 Hz, JHP ¼ 22.3 Hz, 2H,
CH(OH), major diastereomer; 31P{1H}, 5.60 (pseudo d,
3JHH ¼ 6.5 Hz)), 5.44 (ddd, 1H, 3JHH ¼ 3.0 Hz, 3JHH ¼ 9.3 Hz,
2JHP ¼ 12.0 Hz, 2H, CH(OH), minor diastereomer; 1H{31P}, 5.44 (dd,
3JHH ¼ 3.0 Hz, 3JHH ¼ 9.3 Hz)), 3.50e2.85 (m, overlap of 4H, CH(OH)
CH2P and 4H, (CH3)2CHP (both diastereomers); 1H{31P}, same),
1.60e1.40 (m, 12H, (CH3)2CHP, (both diastereomers); 1H{31P},
D2O with a CDCl3/TMS insert):
d 35.59 (s, minor diastereomer),
32.72 (s, major diastereomer). 13C{1H} NMR (100 MHz, D2O with
a CDCl3/TMS insert, complex coupling of several carbon atoms
results from coupling to two magnetically inequivalent phosphorus
same). 31P{1H} NMR (161 MHz, CD3OD):
d
36.81 (s, minor diaste-
atoms in the structure): d 58.10 (m, CH(OH), minor diastereomer),
reomer), 34.54 (s, major diastereomer). 13C{1H} NMR (100 MHz,
CD3OD, signals of carbon atoms appear as multiplets with
complex splitting patterns that arise from coupling to two
magnetically inequivalent phosphorus atoms in the structure):
57.85 (m, CH(OH), major diastereomer), 19.54 (m, CH2P, minor
diastereomer), 18.35 (m, CH2P, major diastereomer), 12.50e9.12 (m,
CH3CH2P, overlap of diastereomer), 5.42e3.72 (m, CH3CH2P, overlap
of diastereomer). Anal. Calcd for C12H28P2O2Br2: C, 33.82; H, 6.62.
Found: C, 33.92; H, 6.38.
d
58.61 (m, CH(OH), minor diastereomer), 57.69 (m, CH(OH), major
diastereomer), 22.93 (m, CH2P, major diastereomer), 22.93 (m,
CH2P, minor diastereomer), 21.23 (m, CH2P, major diastereomer),
4.3. Procedure for the preparation of the dimer 2c
2
21.62 (d, JCP ¼ 21.8 Hz, C(CH3)2P, minor diastereomer) 19.71 (d,
2JCP ¼ 40.5 Hz, C(CH3)2P, major diastereomer), 16.45e15.35 (m,
overlapping peaks of isopropyl methyl groups, both
In an Ar glovebox, potassium hydride (0.521 g; 13.0 mmol) was
partially dissolved in 13 mL of THF. On a Schlenk line, di(p-tolyl)
phosphine (2.31 g; 10.8 mmol) was slowly added to the mixture.
The color of the solution changed to red-orange and H2 gas was
evolved. The solution was stirred for about 30 min until no more
hydrogen generation was observed. The reaction mixture was
then cooled to ꢂ78 ꢀC and bromoacetadehyde diethyl acetal
(2.12 g; 10.8 mmol) was added over
a 20 min period. The
temperature was brought to 25 ꢀC and 2.5 g of 48% HBr
(14.8 mmol) was added. The mixture was heated at 45 ꢀC for 2 h,
and left in the freezer for 5 h. The precipitate was filtered off in the
air and washed twice with 7 mL of cold water, as well as 15 mL of
a 1:1 mixture of cyclohexanol:ethyl acetate. The precipitate was
then recrystallized by slow diffusion of ether into in of a solution
of 2c in 1:1 methanol:toluene. The purified solid was dried under
high vacuum. Yield: 2.70 g, 87.8%. Anal. Calcd for [C32H36P2O2]
[Br]2[CH3OH][H2O]: C, 54.71; H, 5.84. Found: C, 54.65; H, 6.04. The
diastereomeric ratio was found to be 3:4, as determined by 1H
NMR, and 31P NMR.
4.3.1. Diastereomer 1. major
3
1H NMR (400 MHz, CD3OD):
d
8.06 (dd, JHH ¼ 8.2 Hz,
3JHP ¼ 12.3 Hz, 4H, ArH), 7.63e7.58 (m, 8H, ArH), 7.48 (dd,
3
3
3JHH ¼ 8.2 Hz, JHP ¼ 2.5 Hz, 4H, ArH), 6.19 (dd, JHH ¼ 6.6 Hz,
3JHP ¼ 21.5, 2H, PCH(OH)), 4.37e4.15 (m, 2H, PCH(OH)CH2),
3.99e3.81 (m, 2H, PCH(OH)CH2), 2.53 (s, 6H, CH3), 2.44 (s, 6H, CH3).
Fig. 2. ORTEP views of the molecular structure of 6.Thermal ellipsoids are drawn at
30% probability. Some hydrogen atoms and solvent molecules are omitted for clarity.
Selected interatomic distances [Å] and angles [o]: 6 P1eC1 1.870(3), P1eC4 1.799(3),
P1eC5 1.782(3), C1eC2 1.519(4), C2eC3 1.523(4), C1eO1 1.410(4); C1eP1eC4 97.2(1),
C1eP1eC5 111.8 (1), C1eC2eC3 108.1(2).
1H{31P} NMR (400 MHz, CD3OD):
d
8.06 (d, 3JH-H ¼ 8.2 Hz, 4H, ArH),
3
7.63e7.58 (m, 8H, ArH), 7.48 (d, JHH ¼ 8.2 Hz, 4H, ArH), 6.19 (d,

A.A. Mikhailine et al. / Journal of Organometallic Chemistry 695 (2010) 1824e1830
1829
3
3JHH ¼ 6.6 Hz, 2H, PCH(OH)), 4.18 (dd, JHH ¼ 6.9, 16.3 Hz, 2H, PCH
4.5. Procedure for preparation of the monomer 6
3
(OH)CH2), 3.93 (dd, JHH ¼ 2.2, 16.3 Hz, 2H, PCH(OH)CH2), 2.53 (s,
6H, CH3), 2.44 (s, 6H, CH3).31P{1H} NMR (161 MHz, CD3OD):
d
11.12
Diphenylphosphine (0.500 g, 2.69 mmol) was added to
a mixture of partially dissolved potassium hydride (0.129 g,
(s). 13C{1H} NMR (100 MHz, CD3OD):
d
146.9e146.8 (m, p-C of ArP),
146.7 (m, p-C of ArP), 133.7e133.6 (m, o-CH of ArP), 133.1e133.0 (m,
o-C of ArP), 130.7e130.5 (m, m-C of ArP), 130.4e130.3 (m, m-C of
ArP), 114.0e113.8 (m, ipso-C of ArP), 113.0e112.9 (m, ipso-C of ArP),
61.5e60.7 (m, CH(OH)), 22.3e21.6 (m, CH2CH(OH)), 20.5 (s, CH3),
20.3 (s, CH3).
3.22 mmol) in 15 mL of THF. The color of the solution changed to
red-orange and H2 gas evolution was observed. The solution was
stirred until no more hydrogen generation was observed. The
reaction mixture was then cooled to 0 ꢀC and ClCH2CH2CH2CH
(OCH2CH2O) (0.404 g, 2.69 mmol) was added over a 20 min period.
The mixture was warmed to 25 ꢀC and then 10 mL of 5 M HCl was
added. The mixture was heated at 45 ꢀC overnight. The solvent was
completely evaporated under vacuum to give a yellow-white solid.
The solid was redissolved in a minimum amount of MeOH and
diethyl ether was slowly diffused into the solution. White crystals
were collected by quick filtration in air and taken into a glovebox
where the recrystallization was repeated to give the analytically
4.3.2. Diastereomer 2. minor
3
1H NMR (400 MHz, CD3OD):
d
7.96 (dd, JHH ¼ 8.3 Hz,
3JHP ¼ 12.5 Hz, 4H, ArH), 7.86 (dd, 3JHH ¼ 8.3 Hz, 3JHH ¼ 12.0 Hz, 4H,
3
3
ArH), 7.63e7.58 (m, 4H, ArH), 7.56 (dd, JHH ¼ 8.1 Hz, JHP ¼ 3.2 Hz,
4H, ArH), 5.81 (ddd, 3JHH ¼ 2.7, 9.3 Hz, JHP ¼ 16.3 Hz, 2H, PCH(OH)),
4.37e4.15 (m, 2H, PCH(OH)CH2), 3.99e3.81 (m, 2H, PCH(OH)CH2),
2.50 (s, 6H, CH3), 2.49 (s, 6H, CH3). 1H{31P} NMR (400 MHz, CD3OD):
pure product. Yield: 0.645 g, 82%. 1H NMR (400 MHz, CD3OD):
3
3
3
d
7.96 (d, JHH ¼ 8.3 Hz, 4H, ArH), 7.86 (d, JHH ¼ 8.3 Hz, 4H, ArH),
d
8.02e7.62 (m, 10H, ArH), 5.48 (dt, JHH ¼ 6.0 Hz, JHP ¼ 7.6 Hz 1H,
3
7.63e7.58 (m, 4H, ArH), 7.56 (d, JH-H ¼ 8.1 Hz, 4H, ArH), 5.81 (dd,
CH(OH)), 3.22e2.99 (m, 2H, CH2CH(OH)), 2.59e2.12 (m, 4H, PCH2
3JHH ¼ 2.7, 9.3 Hz, 2H, PCH(OH)), 4.27 (dd, JHH ¼ 9.3, 16.2 Hz, 2H,
and PCH2CH2). 1H{31P} NMR (400 MHz, CD3OD):):
d
8.02e7.62 (m,
3
3
3
PCH(OH)CH2), 3.83 (dd, JHH ¼ 2.7, 16.2 Hz, 2H, PCH(OH)CH2), 2.50
10H, ArH), 5.48 (t, JHH ¼ 6.0 Hz, 1H, CH(OH)), 3.22e2.99 (m, 2H,
(s, 6H, CH3), 2.49 (s, 6H, CH3). 31P{1H} NMR (161 MHz, CD3OD):
CH2CH(OH)), 2.59e2.12 (m, 4H, PCH2 and PCH2CH2). 31P{1H} NMR
d
16.06 (s). 13C{1H} NMR (100 MHz, CD3OD):
d
147.5 (m, p-C of ArP),
(161 MHz, CD3OD):
134.7 (d, JCP ¼ 3.0 Hz, p-C of PhP), 134.4(d, JCP ¼ 3.2 Hz, p-C of
d
38.0 (s). 13C{1H} NMR (100 MHz, CD3OD):
147.0e146.9 (m, p-C of ArP), 133.4e133.3 (m, m-C of ArP),
132.9e132.8 (m, m-C of ArP), 131.4e131.3 (m, o-C of ArP),
130.5e130.4 (m, o-C of ArP), 112.8e112.3 (m, ipso-C of ArP),
111.9e111.5 (m, ipso-C of ArP), 62.2 (pdd, CH(OH)), 23.6 (pdd,
CH2CH(OH)), 20.4 (s, CH3).
d
Ph’P), 133.7 (d, JCP ¼ 9.0 Hz, o-C of PhP), 132.8 (d, JCP ¼ 9.0 Hz, o-C of
Ph’P), 130.1 (d, JCP ¼ 11.8 Hz, m-C of PhP),129.4 (d, JCP ¼ 12.4 Hz, m-C
of PhP), 118.4 (d, JCP ¼ 73.3 Hz, ipso-C of PhP), 115.5 (d, JCP ¼ 78.2 Hz,
ipso-C of Ph’P), 72.5 (d, JCP ¼ 55.9 Hz, C(OH)), 34.5 (d, JCP ¼ 18.0 Hz,
CH2CH(OH)), 21.3 (d, JCP ¼ 38.6 Hz, PCH2), 21.4 (d, JCP ¼ 17.2 Hz,
PCH2CH2). Anal. Calcd for [C16H18POCl][ MeOH]: C, 62.80; H, 6.83.
Found: C, 62.33; H, 6.18. Crystals suitable for the X-ray diffraction
experiment were obtained by slow diffusion of the diethyl ether
into a saturated solution of 6 in methanol.
4.4. Procedure for preparation of the tetramer 4
The synthetic procedure for the preparation of 2-(2-chlor-
oethyl)-1, 3-dioxolane was used as described [13] without major
changes. Potassium hydride (3.86 g; 96.3 mmol) was partially dis-
solved in 25 mL of THF in an Ar glovebox. At a Schlenk line under Ar,
diphenylphosphine (14.9 g; 80.0 mmol) was slowly added to the
mixture. The color of the solution changed to red-orange and H2 gas
evolved. The solution was stirred until no more hydrogen genera-
tion was observed. The reaction mixture was then cooled to 0 ꢀC
and the 2-(2-chloroethyl)-1, 3-dioxolane (10.9 g; 80.2 mmol) was
added over a 20 min period. The mixture was warmed to 25 ꢀC and
20 mL of 5 M HCl was added. The mixture was heated at 45 ꢀC over
night. The volume of the solution was reduced to 2/3 of the original
volume to give a milky white solution that was left in the cooler
(Te0 to ꢂ5 ꢀC) overnight to give a white precipitate. The precipitate
was filtered and washed twice with 7 mL of cold H2O and diethyl
ether (10 mL). The precipitate was then recrystallized from a satu-
rated solution in MeOH by slow diffusion of diethyl ether. White
rhombic crystals were filtered and washed with diethyl ether and
pentanes to give the analytically pure product as the trihydrate.
4.6. Structure determination of 2a, 2c and 6 by X-ray diffraction
X-ray crystallographic data were collected on a BrukereNonius
Kappa-CCD diffractometer with use of monochromated MoKa
radiation (
l
¼ 0.71073 Å) and were measured with a combination of
4
scans and
u
scans with k offsets, to fill the Ewald sphere.
Table 2
Summary of crystal data and details of intensity collection and least-squares
refinement parameters for 2a, 2c and 6.
2a
2c
6
Empirical
formula
Fw
C18H44Br2O4P2
C32H36Br2O2P2
C17H22ClO2P
546.29
150(1)
674.37
273(1)
324.77
150(1)
T [K]
Yield: 19.6 g, 88%. 1H{31P} NMR (400 MHz, CD3OD):
d 7.95e7.55 (m,
Space group
A [Å]
B [Å]
P
P
P21
ꢂ1
ꢂ1
40H, ArH), 6.60 (dd, 3JHH ¼ 11.2 Hz, 4H, PCH(OH)), 4.41e4.23 (m, 4H,
CH2CH (OH)), 3.03e2.84 (m, 4H, CH2CH (OH)), 2.80e2.63 (m, 4H,
PCH2CH2),1.30e1.22 (m, 4H, PCH2CH2). 1H NMR (400 MHz, CD3OD):
7.95e7.55 (m, 40H, ArH), 6.60 (pd (pseudo doublet), 3JHH ¼ 10.4 Hz,
4H, PCH(OH)), 4.41e4.23 (pq, 4H, CH2CH (OH)), 3.03e2.84 (pq, 4H,
CH2CH (OH)), 2.80e2.63 (pt, 4H, PCH2CH2), 1.30e1.22 (bs, 4H,
6.9663(5)
9.7835(4)
10.4298(7)
74.240(4)
71.618(3)
74.063(4)
635.17(7)
1
9.1690(9)
10.4027(11)
11.3044(14)
106.400
108.023
108.758
879.25(17)
1
12.5293(3)
15.4572(8)
8.8188(7)
90.00
100.445(3)
90.00
1679.62(16)
4
1.284
C [Å]
a
b
g
[ꢀ]
[ꢀ]
[ꢀ]
V [Å3]
PCH2CH2).
d
31P{1H} NMR (161 MHz, CD3OD):
d
27.08 (s). 13C{1H}
Z
P [mgcmꢂ3
Reflections
]
1.428
6621
1.274
3817
NMR (100 MHz, CD3OD):
d
135.0 (s, p-C of PhP), 134.8 (s, p-C of
13 223
Ph’P), 134.0 (m, m-C of PhP), 133.4 (m, m-C of Ph’P), 129.9 (m, o-C of
PhP), 129.5 (m, o-C of Ph’P), 115.7 (d, JCP ¼ 85 Hz, ipso-C of PhP),
114.7 (d, JCP ¼ 84 Hz, ipso-C of Ph’P), 63.1 (m, PCH(OH)), 23.7 (m,
PCH2CH2CH(OH)), 16.5 (m, PCH2CH2CH(OH)). Anal. Calcd for
[C60H64P4O4Cl4][2MeOH][2H2O] C, 61.29; H, 6.30. Found: C, 61.23;
H, 5.84.
collected
Unique
reflections
Data;
parameters
R1; wR2
2862
3817
3832
[R(int) ¼ 0.0436]
[R(int) ¼ 0.060]
[R(int) ¼ 0.0975]
2862; 128
3817; 175
3832; 208
0.0356; 0.087
0.0554; 0.1532
0.0556; 0.1392

1830
A.A. Mikhailine et al. / Journal of Organometallic Chemistry 695 (2010) 1824e1830
(e) M. Ito, A. Osaku, C. Kobayashi, A. Shiibashi, T. Ikariya, Organometallics
Crystals of the tetramer 4 were obtained by slow diffusion of
28 (2009) 390e393;
diethyl ether into a saturated solution of 4 in methanol. The pres-
ence of large cavities in the lattice that originate from the tetra-
meric nature of the phosphonium ion allowed co-crystallization of
a large number of solvent molecules (water and methanol as
determined from 1H NMR studies) that appeared to be highly
disordered. Attempts to model the solvent molecules were not
successful. If the SQUEEZE option in PLATON was used to minimize
the overall level of the disorder in the lattice, then the density of the
crystal was found to be lowered by a significant amount because of
the formation of large cavities. As a result, we cannot report the
crystal structure of 4 in this manuscript. On the other hand, this
preliminary X-ray analysis result confirms the proposed structure
with a high level of confidence. Table 2.
(f) I.D. Kostas, Cur. Org. Synth 5 (2008) 227e249;
(g) P.-H. Leung, Acc. Chem. Res. 37 (2004) 169e177;
(h) L. Boubekeur, L. Ricard, N. Mezailles, M. Demange, A. Auffrant, P. Le
Floch, Organometallics 25 (2006) 3091e3094;
(i) D.B.G. Williams, M. Pretorius, J. Mol. Cat. A: Chem. 284 (2008) 77e84.
[3] (a) B. Demerseman, Eur. J. Inorg. Chem. (2001) 2347e2359;
(b) P. Braunstein, Chem. Rev. 106 (2006) 134e159;
(c) P. Kuhn, D. Semeril, D. Matt, M.J. Chetcuti, P. Lutz, Dalton Trans. (2007)
515e528;
(d) M. Agostinho, V. Rosa, T. Aviles, R. Welter, P. Braunstein, Dalton Trans.
(2009) 814e822;
(e) E. Lindner, S. Pautz, M. Haustein, Coord. Chem. Rev. 155 (1996)
145e162.
[4] (a) D. Matt, R. Ziessel, A.D. Cian, J. Fischer, New J. Chem. 20 (1996) 1257e1263;
(b) M.P. Sokolov, K. Issleib, Z.Chem. 17 (1977) 365e366;
(c) H. Brisset, Y. Gourdel, P. Pellon, M. Le Corre, Tetrahedron Lett. 34 (1993)
4523e4526;
(d) G.D. Vaughn, J.A. Gladysz, J. Am. Chem. Soc. 108 (1986) 1473e1480;
(e) D.V. Moiseev, B.O. Patrick, B.R. James, Inorg. Chem. 46 (2007)
11467e11474.(f) D.V. Moiseev, P. Marcazzan, B.R. James, Can. J. Chem. 87
(2009) 582e590.
Acknowledgment
R. H. M. thanks NSERC Canada for a Discovery Grant.
[5] (a) V. Rautenstrauch, X. Hoang-Cong, R. Churlaud, K. Abdur-Rashid, R.
H. Morris, Chem. Eur. J 9 (2003) 4954e4967;
Appendix A. Supplementary material
(b) J.R. Dilworth, S.D. Howe, A.J. Hutson, J.R. Miller, J. Silver, R.
M. Thompson, M. Harman, M.B. Hursthouse, J. Chem. Soc., Dalton Trans.
(1994) 3553e3562;
(c) R. El Mail, M.A. Garralda, R. Hernandez, L. Ibarlucea, E. Pinilla, M.
R. Torres, M. Zarandona, Eur J. Inorg. Chem. (2005) 1671e1677;
(d) J.-X. Gao, H. Zhang, X.-D. Yi, P.-P. Xu, C.-L. Tang, H.-L. Wan, K.-R. Tsai,
T. Ikariya, Chirality 12 (2000) 383e388.
Data collection parameters and crystallographic information can
be obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif. (2a CCDC
749473, 2c CCDC 749474, 6 CCDC 749475).
[6] Y.-Y. Li, Z.-R. Dong, J.-X. Gao, Current Topics in Catalysis 6 (2007) 35e54.
[7] A.A. Mikhailine, E. Kim, K. Dingels, A.J. Lough, R.H. Morris, Inorg. Chem. 47
(2008) 6587e6589.
References
[8] A.A. Mikhailine, A.J. Lough, R.H. Morris, J. Am. Chem. Soc. 131 (2009)
1394e1395.
[9] M. Ranocchiari, A. Mezzetti, Organometallics 28 (2009) 1286e1288.
[10] (a) C.D. Meyer, C.S. Joiner, J.F. Stoddart, Chem. Soc. Rev. 36 (2007) 1705e1723;
(b) N.V. Gerbeleu, V.B. Arion, J. Burgess, Template Synthesis of Macrocyclic
Compounds. Wiley-VCH, Weinheim, 1999.
[11] A. Bianchi, K. Bowman-James, E. García-España, Supramolecular Chemistry of
Anions. Wiley VCH, Weinheim, Germany, 1997.
[12] I. Alfonso, M. Bolte, M. Bru, M.I. Burguete, S.V. Luis, J. Rubio, J. Am. Chem. Soc.
130 (2008) 6137e6144.
[1] (a) R.A. Baber, M.F. Haddow, A.J. Middleton, A.G. Orpen, P.G. Pringle, A. Haynes,
G.L. Williams, R. Papp, Organometallics 26 (2007) 713e725;
(b) A.J. Poe, Dalton Trans. (2009) 1999e2003;
(c) C.A. Tolman, Chem. Rev. 77 (1977) 313e348.
[2] (a) P. J. Guiry, C.P. Saunders, Adv. Syn. Cat 346 (2004) 497e537;
(b) D. Benito-Garagorri, K. Kirchner, Acc. Chem. Res. 41 (2008) 201e213;
(c) R.J. Detz, S.A. Heras, R. de Gelder, P. van Leeuwen, H. Hiemstra, J.N.
H. Reek, J.H. van Maarseveen, Organic Lett. 8 (2006) 3227e3230;
(d) S.E. Durran, M.B. Smith, S.H. Dale, S.J. Coles, M.B. Hursthouse, M.E. Light,
Inorg. Chim. Acta 359 (2006) 2980e2988;
[13] P.B. Shrestha-Dawadi, J. Lugtenburg, Eur. J. Org. Chem. 23 (2003) 4654e4663.