Efficient Multigram Syntheses of Air-Stable, Fluorescent Primary Phosphines via Palladium-Catalyzed Phosphonylation of Aryl Bromides

Article abstract of DOI:10.1055/s-0034-1378336

Air-stable, fluorescent primary phosphines have been prepared on a multigram scale. The two key synthetic steps are an optimized palladium-catalyzed phosphonylation of aryl bromides and a boron-carbon bond formation reaction. The method provides a valuable synthetic route to novel fluorescent derivatives via a key phosphonate intermediate.

Full text of DOI:10.1055/s-0034-1378336

2622▌
PAPER
paper
Efficient Multigram Syntheses of Air-Stable, Fluorescent Primary Phosphines
via Palladium-Catalyzed Phosphonylation of Aryl Bromides
Syntheses of Air-Stable, Fluorescent Primary Phosphines
Laura H. Davies, Jennifer F. Wallis, Michael R. Probert, Lee J. Higham*
School of Chemistry, Bedson Building, Newcastle University, Newcastle upon Tyne, NE1 7RU, UK
Fax +44(191)2226929; E-mail: lee.higham@ncl.ac.uk
Received: 19.12.2013; Accepted after revision: 25.05.2014
Ph2P
PPh2
PH2
PPh2
Abstract: Air-stable, fluorescent primary phosphines have been
prepared on a multigram scale. The two key synthetic steps are an
optimized palladium-catalyzed phosphonylation of aryl bromides
and a boron–carbon bond formation reaction. The method provides
a valuable synthetic route to novel fluorescent derivatives via a key
phosphonate intermediate.
P
Ph
P
PPh2
Triphos
N
N
Key words: phosphorus, palladium, fluorescence, air stability,
Bodipy
N
N
B
B
R
R
Me Me
1a R = Ph
1b R = Me
2
Primary phosphines are largely regarded as troublesome
compounds from the perspective of synthetic methodolo-
gy, due to their reputation as highly air-sensitive, volatile,
toxic, and pyrophoric compounds.1 The low air stability of
many primary phosphines has made them an under-uti-
lized class of ligand; however, they are versatile starting
materials due to functionalization of the phosphorus–hy-
drogen bonds. In recent years, a very limited number of
‘user-friendly’ primary phosphines have been reported,2
some of which take advantage of steric encumbrance in
order to afford kinetic stability towards air oxidation.3 A
handful of other primary phosphines whose unexplained
and surprising air stability cannot be accounted for on ste-
ric grounds have also appeared in the literature.4 Very re-
cently, we discovered the first air-stable fluorescent
primary phosphines 1a and 1b based on Bodipy (Figure
1).5 According to our density functional theory-based
model, their air stability is attributable to the high level of
π-conjugation in the Bodipy backbone.5,6 We have also
shown that despite this resistance to air oxidation (in air,
no oxidation in the solid state or in chloroform solution af-
ter seven days was observed5,6), the phosphino group re-
mains readily transformable to yield fluorescent
derivatives of, for example, Triphos (Figure 1).5 Triden-
tate phosphine 2 has been shown by us to form stable che-
lates with rhenium,5 a metal used as a nonradioactive
substitute for the gamma emitter 99mTc. Such complexes
have potential as dual imaging agents by combining the
desirable photophysical properties of the Bodipy fluoro-
phore7 for in vitro fluorescence microscopy, together with
the in vivo radioimaging afforded by virtue of the 99mTc
center.5
Figure 1 Bodipy primary phosphines 1a and 1b and the tridentate
phosphine 2, which is a fluorescent analogue of Triphos
The potential of 1a and 1b as ligand precursors is thus
clear and we therefore sought an efficient large-scale syn-
thesis of them. The aforementioned synthetic approach
was reevaluated and two problematic steps were identi-
fied as (i) the substitution of the Bodipy fluorides for alkyl
or aryl groups, as the dyes are formed in low yields
(Scheme 1, step ii 44%: 4a, 40%: 4b) and (ii) the carbon–
phosphorus coupling reaction, affording the diethyl phos-
phonates, is hindered by a reaction time of three days and
a yield of 53% for 5b (Scheme 1, step iii). These factors
limited the previous syntheses to a scale of less than one
gram.5
Herein we report a significantly improved synthetic meth-
od for preparing the primary phosphines 1a and 1b in mul-
tigram quantities via a simplified and improved approach,
with the focus on the phosphonylation and the boron–
carbon bond formation steps. We also include a number of
new solid-state structures, together with a previously un-
reported, highly fluorescent Bodipy phosphonate.
The syntheses of the fluorescent primary phosphines 1a
and 1b start from the preparation of the Bodipy aryl bro-
mide derivative 3 in a one-pot condensation reaction
between the commercially available reagents 4-bromo-
benzaldehyde and 3-ethyl-2,4-dimethyl-1-pyrrole, in a
moderate yield of 47%, following the literature procedure
(Scheme 1, step i).8 Bodipy formation reactions are noto-
riously low yielding.7 However, the situation could be im-
proved to 55% by (i) ensuring the dichloromethane and
reagents N-diisopropylethylamine and boron trifluoride
were anhydrous, and (ii) by reducing the amount of tri-
fluoroacetic acid (TFA) added to 0.025 mL, and using sig-
nificantly less N-diisopropylethylamine (6 equiv instead
of 11.4 equiv), boron trifluoride (8 equiv instead of 16
equiv),9 and dichloromethane (100 mL per 5.4 mmol of 4-
bromobenzaldehyde instead of 400 mL per 5.4 mmol).
SYNTHESIS 2014, 46, 2622–2628
Advanced online publication: 24.07.2014
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
DOI: 10.1055/s-0034-1378336; Art ID: ss-2013-n0824-op
© Georg Thieme Verlag Stuttgart · New York

PAPER
Syntheses of Air-Stable, Fluorescent Primary Phosphines
2623
Br
original route with limitations
on yields and reaction times
O
O
O
N
N
P
PH2
Br
B
R
R
(iii)
(ii)
4a R = Ph
4b R = Me
Br
H
N
(i)
(iv)
2
N
N
N
N
N
N
O
B
B
B
O
O
F
F
R
R
R R
P
O
H
3
5a R = Ph
5b R = Me
(v) Δ, or
1a R = Ph
1b R = Me
(vii)
(vi) μ-wave
revised and improved route
via a reaction order change
N
N
B
F
F
6
Scheme 1 Synthesis of 1a and 1b. Reagents and conditions: (i) TFA, DDQ, i-Pr2NEt, Et2O·BF3, CH2Cl2, r.t.; (ii) PhLi or MeLi, THF, r.t.; (iii)
i-Pr2NEt, HP(O)(OEt)2, DPPB, and [Pd(OAc)2]/DMSO (4a) or [Pd(dba)2]/toluene (4b), 90 °C; (iv) LiAlH4, Me3SiCl, THF, –78 °C to r.t.; (v)
i-Pr2NEt, HP(O)(OEt)2, [Pd(PPh3)4], DMSO, 90 °C; (vi) μ-wave, DMSO, 120 °C; (vii) PhMgBr or MeMgBr, THF, r.t.
These changes resulted in a simplified work-up, higher Originally, access to the phosphonate derivatives 5a and
yields, and a lower reaction expense.
5b came via the aryl bromides 4a and 4b; however, the
boron–carbon and phosphorus–carbon bond formation
steps (ii and iii, Scheme 1) were low yielding, for 4b and
5b in particular (40% and 53%, respectively).5 Out of the
two primary phosphines, 1b has the superior quantum
yield (ΦF = 1a 0.042; 1b 0.33).5 Therefore, the methyla-
tion reaction of step ii was carried out by using methyl-
magnesium bromide instead of methyllithium, but no
improvements were observed. Compound 4b also suf-
fered from low solubility in DMSO, which therefore re-
quired a different catalyst/solvent combination to be
found (iii, Scheme 1). A revised strategy was then consid-
ered in order to overcome this limiting factor. A phospho-
nylation reaction of 3 ought to give the novel phosphonate
6, which could then in principle be arylated and alkylated
to give the target compounds 5a and 5b, respectively
(Scheme 1, lower route). We succeeded in effecting the
transformation of aryl bromide 3 into the F-Bodipy phos-
phonate 6 via a palladium-catalyzed phosphonylation
(Scheme 1, step v). Initially the same protocol was fol-
lowed as for the synthesis of 5a (step iii), using the cata-
lyst palladium(II) acetate with the ligand 1,4-
bis(diphenylphosphinobutane) (DPPB); however, only a
yield of 58% was achieved. A number of protocols de-
scribing the transition metal-catalyzed phosphonylation
reaction and mechanistic studies of phosphorus–carbon
bond formation have been published, which show that the
palladium source and phosphine ligand can both affect the
reaction yields.10,11 Following on from our earlier prelim-
inary investigations, optimization of the reaction condi-
tions was carried out by screening the phosphonylation of
The difluoro-substituted aryl bromide 3 was characterized
by X-ray crystallography and its solid-state structure is
depicted in Figure 2.
Figure 2 View of 3 with 50% probability displacement ellipsoids.
Hydrogen atoms have been omitted for clarity (see Supporting Infor-
mation for crystallographic details).
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 2622–2628

2624
L. H. Davies et al.
PAPER
3 to obtain 6 (Table 1). The amount of starting material The effect of using microwave (MW) irradiation for our
(0.54 mmol), the relative ratio of substrates, the reaction phosphonylation reactions was tested. The conversion of
time (42 h), and solvent (10 mL of anhydrous DMSO) phenyl bromide to diethyl phenylphosphonate has been
were kept constant throughout. Only a small excess of di- shown to be accelerated considerably by using a micro-
ethyl phosphite (1.2 equiv) was used, because it has been wave power source.12 The same reaction parameters from
reported that a significant excess of this compound can our initial thermal screening were chosen (vide supra), but
deactivate the palladium catalyst.11
using closed microwave vessels, at 120 °C for 10 minutes
(Table 1, entry 9). The yield was very slightly decreased
to 82% but the reaction time was dramatically reduced
from 42 hours to 10 minutes. Following reports of decom-
position of DMSO and starting materials at high concen-
tration in microwave irradiation,12,13 THF was also tried as
an alternative solvent, but this only gave a yield of 64%.
First, the phosphine ligand that was added to the palladi-
um(II) acetate was varied and quickly it was found that
this had a significant impact on the reaction yield (Table
1, entries 1–5). 1,3-Bis(diphenylphosphinopropane)
(DPPP, entry 3) and triphenylphosphine (entry 5) were
found to be the best ligands (78% and 80%, respectively).
Next the palladium source was changed, which showed An X-ray crystal structure of 6 was obtained from the sin-
that yields of 80% could also be achieved by employing gle crystal obtained by slow evaporation of a dichloro-
bis(dibenzylideneacetone)palladium(0) and the ligand methane solution (Figure 3). The P–C bond length of
bis(diphenylphosphino)ferrocene (DPPF, entry 7). How- 1.7868(18) Å correlates well with other reported aryl-
ever, the best yield achieved was 85%, using phosphonate structures13b and the bond lengths in the di-
tetrakis(triphenylphosphine)palladium(0) as catalyst (en- azaindacene core are similar to that found for other F-
try 9). This reaction was also attempted using lower cata- Bodipy derivatives.14 The novel F-Bodipy phosphonate 6
lytic loadings (5 mol%, entry 10); however, this reduced shows excellent photophysical properties.
the yield to 50% and thus 10 mol% was required for this
transformation to obtain an optimum yield.
Table 1 Thermal and Microwave Reaction Conditions for the Palla-
dium-Catalyzed Phosphonylation of Aryl Bromide 3a
O
O
O
Br
P
HP(O)(OEt)2
i-Pr2NEt
[Pd], ligand
DMSO or THF
90 °C or 120 °C
N
N
N
N
B
B
F
F
F
F
3
6
Entry
Palladium catalyst
Ligand
Yield (%)b
1
2
Pd(OAc)2 (10 mol%)
Pd(OAc)2 (10 mol%)
Pd(OAc)2 (10 mol%)
Pd(OAc)2 (10 mol%)
Pd(OAc)2 (10 mol%)
Pd(dba)2 (10 mol%)
Pd(dba)2 (10 mol%)
Pd(dba)2 (10 mol%)
Pd(PPh3)4 (10 mol%)
Pd(PPh3)4 (5 mol%)
DPPB
DPPE
DPPP
DPPF
PPh3
DPPP
DPPF
PPh3
–
58
10
Figure 3 View of 6 with 50% probability displacement ellipsoids.
Hydrogen atoms have been omitted for clarity (see the Supporting In-
formation for crystallographic details).
3
78
4
71
5
80
The absorption maximum is at 526 nm (in THF), which is
assigned to the S0-S1 (π-π*) electronic transition associat-
ed with the Bodipy core. The high molar absorption coef-
ficient (ε = 56 000 M–1cm–1) is in keeping with previous
findings for Bodipy compounds.7 Room temperature flu-
orescence is readily detected with an emission maximum
of 540 nm (in THF). The fluorescence quantum yield (ΦF)
of 0.56 is high and only somewhat reduced from the aryl
bromide 3 (ΦF = 0.65).15 The photophysical data and spec-
tra can be found in the Supporting Information. Phospho-
nates have received increasing attention for imaging bone
diseases, due to their high calcium(II) affinity, which re-
sults in accumulation in areas of increased bone metabo-
lism.16 Thus this highly fluorescent phosphonate may
6
28
7
80
8
76
9
85 (82c, 64d)
50
10
–
a Conditions: 3 (250 mg, 0.54 mmol), i-Pr2NEt (0.29 mL), diethyl
phosphite (0.08 mL), bidentate ligand (0.054 mmol) or PPh3 (0.108
mmol), anhydrous DMSO (10 mL), 90 °C, 42 h.
b Isolated yields after workup and column chromatography.
c Microwave, closed vessel: DMSO (10 mL), 120 °C, 10 min.
d Microwave, closed vessel: THF (10 mL), 120 °C, 10 min.
Synthesis 2014, 46, 2622–2628
© Georg Thieme Verlag Stuttgart · New York

PAPER
Syntheses of Air-Stable, Fluorescent Primary Phosphines
2625
have applications in the study of bone cells by fluores- steps ii and iii in Scheme 1 is important. Novel phospho-
cence microscopy.
nate 6 can now by synthesized in high yield by an opti-
mized palladium-catalyzed phosphonylation of Bodipy
aryl bromide 3 by conventional thermal heating or by MW
irradiation, whilst phosphonates 5a and 5b can now be
synthesized from an arylation or alkylation of 6 using
Grignard reagents, also in high yields (>79%). For the
highly fluorescent primary phosphine 1b, over the last
three steps there is now an overall yield of 61%, which is
greatly improved compared to the original 20%. All reac-
tions have been carried out on a multigram scale without
reductions in yields.
Gratifyingly, we then discovered that the subsequent ary-
lation/alkylation of F-Bodipy phosphonate 6 to the C-
Bodipy phosphonates 5a and 5b respectively, via the ad-
dition of two equivalents of a 3.0 M solution of phenyl or
methylmagnesium bromide, was successful. High yields
of up to 86% were achieved (Scheme 1, step vii) and this
is therefore the superior route.
The final step in the synthesis is the reduction of the phos-
phonates 5a and 5b to the primary phosphines 1a and 1b
using trimethylsilyl chloride and lithium aluminum hy-
dride as co-reductants. The combination of reducing
agents allows for high yields (≥84%).17 The reduction was
found to be concentration dependent, in which the reac-
tion was only successful using very dilute solutions of the
phosphonates (0.20 mmol per 10 mL of THF). In too con-
centrated solutions other primary phosphines were
formed and often none of the desired product (δ = 1a
All air- and/or water-sensitive reactions were performed under a N2
atmosphere using standard Schlenk line techniques. Toluene (sodi-
um), THF (sodium/benzophenone ketyl), and CH2Cl2 (CaH2) were
dried and distilled prior to use. DMSO (Aldrich) was purchased in
an anhydrous state and stored over molecular sieves. All other
chemicals were used as received without further purification. Petro-
leum ether (PE) used refers to the fraction boiling in the 40–60 °C
range. Microwave-assisted reactions were performed in 35 mL
closed vessels on a CEM Discover model under automated power
control based on temperature feedback. Flash chromatography was
performed on silica gel from Fluorochem (silica gel, 40–63 μm, 60
Å, LC301). TLC was performed on Merck aluminum-based plates
with silica gel and fluorescent indicator 254 nm. 1H, 13C{1H},
31P{1H}, 19F{1H}, and 11B{1H} NMR spectra were recorded on a
JEOL Lambda 500 (1H 500.16 MHz) or JEOL ECS-400 (1H 399.78
MHz) spectrometer at r.t. (21 °C); 1H and 13C shifts were relative to
1
−121.5; 1b −121.7. JP,H = 1a 202.5 Hz; 1b 202.5 Hz),
which therefore suggests decomposition of the Bodipy
backbone in the basic conditions.18 The remarkable air-
stability of these primary phosphines allowed for their
straightforward purification on silica gel media, a proce-
dure seldom considered or indeed applicable for this fam-
ily of compound. Indeed, recrystallization of 1a in air
from dichloroethane–methanol allowed us to determine
its solid-state structure by X-ray crystallography (Figure
TMS, 31P relative to 80% H3PO4, 11B relative to Et2O·BF3, and 19
relative to CFCl3 (all shifts given in ppm). Infrared spectra were re-
F
4), which is quite rare for uncoordinated primary phos- corded on a Varian 800 FT-IR Scimitar spectrophotometer
equipped with an ATR sampling accessory. Mass spectrometry was
carried out by the EPSRC National Mass Spectrometry Service
Centre at Swansea. Absorption spectra were recorded with a Hitachi
Model U-3310 spectrophotometer while fluorescence studies were
phines. The P–C bond length of 1.844(3) Å compares well
to the few other primary phosphines which have been
characterized crystallographically.2,4,19
recorded with a Hitachi F-4500 fluorescence spectrophotometer.
Compounds 4a and 4b were prepared according to literature proce-
dures.5 Analytical data for compounds 1a, 1b, 3, 5a, and 5b were
consistent with the previously published values.5
X-ray Crystallography
All data were collected on an Oxford Diffraction (now Agilent
Technologies) Gemini A Ultra diffractometer at 150.0(1) K, using
MoKα radiation with λ = 0.71073 Å. Analytical absorption correc-
tions were applied to all datasets. Data were collected and reduced
using the CrysAlisPro software.20 Structures were solved using di-
rect methods and refined on all unique F2 values, using the Olex221
interface to the SHELX22 suite of programs. All non-H atoms were
refined anisotropically. H atoms were placed using geometric riding
constraints and isotropic parameters related to their parent atom (ex-
cept for the H atoms of the PH2 group of 1a, which were refined
with restrained bond lengths and free rotation). Minor disorder was
present in all structures, with final occupancies fixed at their freely
refined values. Key crystallographic data are given in Table 2.23
8-(4-Bromophenyl)-4,4-difluoro-1,3,5,7-tetramethyl-2,6-dieth-
yl-4-bora-3a,4a-diaza-s-indacene (3)
TFA (0.025 mL) was added dropwise to a stirred solution of 3-eth-
Figure 4 View of 1a with 50% probability displacement ellipsoids.
yl-2,4-dimethyl-1H-pyrrole (21.9 mL, 162 mmol) and 4-bromo-
Hydrogen atoms have been omitted for clarity (see the Supporting In-
benzaldehyde (15.0 g, 81 mmol) in anhydrous CH2Cl2 (1500 mL).
formation for crystallographic details).
The reaction mixture was stirred at r.t. under N2 in a darkened flask
overnight. DDQ (20.2 g, 81 mmol) was added in a single portion,
and the mixture was stirred at r.t. for 2 h. Anhydrous i-Pr2NEt (84.7
mL, 486 mmol) and Et2O·BF3 (80.0 mL, 648 mmol) were added,
and the mixture was stirred at r.t. overnight. The reaction mixture
was washed with H2O (2 × 150 mL) and brine (2 × 100 mL). The
In summary, we have optimized a far more economical
and versatile synthetic route to valuable fluorescent pri-
mary phosphines. In particular, the switch of the synthetic
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 2622–2628

2626
L. H. Davies et al.
PAPER
Table 2 Selected Crystallographic Data for 1a, 3, and 6
Compound
1a
3
6
Formula
Crystal system
Space group
a (Å)
C35H38BN2P
monoclinic
P21/c
C23H26BBrF2N2
triclinic
P2
C27H36BF2N2O3P
monoclinic
P21/n
12.1121(7)
23.5024(14)
11.3376(6)
90
7.623(2)
7.700(2)
9.2409(19)
84.47(2)
81.04(2)
83.39(2)
530.6(3)
1
7.4182(3)
12.1548(6)
29.7551(13)
90
b (Å)
c (Å)
α (°)
β (°)
113.684(7)
90
93.611(4)
90
γ (°)
V (Å3)
2955.6(3)
4
2677.6(2)
4
Z
R (F, F2>2σ)
Rw (F2, all data)
CCDC depos. no.
0.069
0.045
0.037
0.174
0.1071
0.119
967539
967537
967538
separated organic layer was dried (MgSO4), filtered, and the solvent
removed in vacuo to yield a dark-violet solid with a green tint. Pu-
rification using column chromatography on silica gel (toluene) af-
forded the intended product. A sample suitable for X-ray
crystallographic analysis was obtained from CH2Cl2; yield: 20.50 g
(55%); dark red/purple solid; mp 239–242 °C; Rf = 0.6 (toluene).
Method B: [Pd(PPh3)4] (0.063 g, 0.054 mmol), and aryl bromide 3
(0.250 g, 0.54 mmol) were placed in a microwave vessel and dis-
solved in anhydrous DMSO (10 mL). i-Pr2NEt (0.29 mL, 1.63
mmol) and diethyl phosphite (0.084 mL, 0.65 mmol) were added
and the solution was purged with N2 for 10 min. The closed vessel
was irradiated with microwaves at 120 °C for 10 min. H2O (10 mL)
was added to the reaction mixture and the suspension was extracted
with CH2Cl2 (20 mL). The organic layer was washed with H2O (1 ×
20 mL) and brine (1 × 20 mL), dried (MgSO4), filtered, and the sol-
vent removed in vacuo to yield a red/purple solid. Purification was
performed by column chromatography on silica gel (EtOAc–PE,
3:2) to give the intended product; yield: 0.230 g (82%); red solid;
mp 152–154 °C; Rf = 0.3 (EtOAc–PE, 3:2).
IR (neat): 2971 (w), 2901 (w), 1532 (s), 1473 (m), 1405 (s), 1316
(w), 1259 (m), 1182 (s), 1065 (m), 973 (s), 754 (s), 623 cm–1 (m).
1H NMR (400 MHz, CDCl3): δ = 7.61 (d, 3JH,H = 8.2 Hz, 2 H), 7.16
(d, 3JH,H = 8.2 Hz, 2 H), 2.52 (s, 6 H), 2.29 (q, 3JH,H = 7.8 Hz, 4 H),
1.30 (s, 6 H), 0.96 (t, 3JH,H = 7.8 Hz, 6 H).
13C{1H} NMR (100 MHz, CDCl3): δ = 154.2, 138.6, 138.2, 134.9,
133.1, 132.4, 130.6, 130.2, 123.1, 17.2, 14.7, 12.6, 12.0.
IR (neat): 2971 (w), 2901 (w), 1534 (m), 1406 (m), 1315 (m), 1250
(m), 1183 (s), 1056 (m), 975 cm–1 (s).
1H NMR (400 MHz, CDCl3): δ = 7.96 (dd, 3JH,H = 8.3 Hz,3JH,P = 12.9
Hz, 2 H), 7.42 (dd, 3JH,H = 8.1 Hz, 4JH,P = 3.9 Hz, 2 H), 4.17 (m, 4
H), 2.52 (s, 6 H), 2.29 (q, 3JH,H = 7.6 Hz, 4 H), 1.34 (t, 3JH,H = 6.9 Hz,
6 H), 1.26 (s, 6 H), 0.97 (t, 3JH,H = 7.6 Hz, 6 H).
19F{1H} NMR (376 MHz, CDCl3): δ = –145.6 [q (equal intensity),
1JF,B = 31.8 Hz, 2 F].
11B{1H} NMR (128 MHz, CDCl3): δ = –0.2 (t, 1JF,B = 31.8 Hz, 1 B).
HRMS (ESI+): m/z [M]+ calcd for C26H26BBrF2N2: 457.1368;
found: 457.1368.
13C{1H} NMR (100 MHz, CDCl3): δ = 154.1, 139.9, 138.3, 137.9,
UV (THF): λmax (ε) = 526 nm (78 000).
132.9, 132.23 (d, JC,P = 10.0 Hz), 130.1, 129.1 (d, 1JC,P = 187.4 Hz),
128.5 (d, JC,P = 15.3 Hz), 62.2 (d, 2JC,P = 5.8 Hz), 16.9, 16.2 (d, 3JCP
=
8-[(4-Diethylphosphonato)phenyl]-4,4-difluoro-1,3,5,7-te-
tramethyl-2,6-diethyl-4-bora-3a,4a-diaza-s-indacene (6)
Method A: [Pd(PPh3)4] (3.0 g, 2.6 mmol), and aryl bromide 3 (12.0
g, 25.9 mmol) were dissolved in anhydrous DMSO (480 mL) under
N2. i-Pr2NEt (13.9 mL, 78.2 mmol) and diethyl phosphite (4.0 mL,
31.2 mmol) were added subsequently and the solution was heated to
90 °C for 42 h. H2O (350 mL) was added to the reaction mixture and
the suspension was extracted with CH2Cl2 (400 mL). The organic
layer was washed with H2O (1 × 150 mL) and brine (1 × 150 mL),
dried (MgSO4), filtered, and the solvent removed in vacuo to yield
a red/purple solid. Purification was performed by column chroma-
tography on silica gel (EtOAc–PE, 3:2) to give the intended prod-
uct. A sample suitable for X-ray crystallographic analysis was
obtained from a CH2Cl2–EtOAc solution; yield: 11.52 g (85%); red
solid.
5.8 Hz), 14.4, 12.3, 11.6.
19F{1H} NMR (376 MHz, CDCl3): δ = –145.6 [q (equal intensity),
1JF,B = 32.0 Hz, 2 F].
31P{1H} NMR (162 MHz, CDCl3): δ = 18.3.
11B{1H} NMR (128 MHz, CDCl3): δ = –0.2 (t, 1JF,B = 32.0 Hz, 1 B).
HRMS (ESI+): m/z [M + H]+ calcd for C27H37BF2N2PO3: 516.2634;
found: 516.2626.
UV (THF): λmax (ε) = 526 nm (56 000).
8-[(4-Diethylphosphonato)phenyl]-4,4-diphenyl-1,3,5,7-te-
tramethyl-2,6-diethyl-4-bora-3a,4a-diaza-s-indacene (5a)
Phosphonate 6 (7.0 g, 13.6 mmol) was dissolved in anhydrous THF
(200 mL). To this solution was added PhMgBr (10.2 mL, 30.6
mmol, 3 M solution in Et2O) dropwise at r.t. The solution was
Synthesis 2014, 46, 2622–2628
© Georg Thieme Verlag Stuttgart · New York

PAPER
Syntheses of Air-Stable, Fluorescent Primary Phosphines
2627
stirred at r.t. until complete consumption of starting material was
observed by TLC (3 h). The reaction was quenched with MeOH (30
mL) and the solvent was evaporated to leave a deep red solid. Puri-
fication was performed by column chromatography on silica gel
(EtOAc–PE, 2:1) to afford the product; yield: 6.78 g (79%); orange
solid; mp 158–160 °C, Rf = 0.4 (EtOAc–PE, 2:1).
ing it in an ice bath. The product was extracted with Et2O (4 × 75
mL), dried (MgSO4), and the solvents were evaporated.
8-[(4-Phosphino)phenyl]-4,4-diphenyl-1,3,5,7-tetramethyl-2,6-
diethyl-4-bora-3a,4a-diaza-s-indacene (1a)
Purification was performed by column chromatography on silica
gel (CH2Cl2–PE, 1:2) to afford the intended product. A sample suit-
able for X-ray crystallographic analysis was obtained from CH2Cl2–
MeOH; yield: 9.06 g (87%); orange solid; mp 207–210 °C; Rf = 0.3
(CH2Cl2–PE, 1:5).
IR (neat): 2960 (w), 1547 (s), 1474 (m), 1386 (s), 1303 (m), 1254
(w), 1168 (m), 1141 (m), 1032 (m), 970 (s), 773 cm–1 (s).
3
1H NMR (400 MHz, CDCl3): δ = 7.98 (dd, JH,H = 7.8 Hz, 3JH,P
=
13.2 Hz, 2 H), 7.53 (dd, 3JH,H = 7.8 Hz, 4JH,P = 4.0 Hz, 2 H), 7.42–
7.40 (m, 4 H), 7.26–7.15 (m, 6 H), 4.18 (m, 4 H), 2.21 (q, 3JH,H = 7.4
Hz, 4 H), 1.79 (s, 6 H), 1.35 (t, 3JH,H = 7.2 Hz, 6 H), 1.30 (s, 6 H),
0.86 (t, 3JH,H = 7.4 Hz, 6 H).
IR (neat): 2963 (w), 2928 (w), 2869 (w), 2285 (w, P–H), 1545 (s),
1469 (m), 1393 (m), 1303 (s), 1169 (s), 1142 (m), 962 (s), 774
cm–1 (s).
1H NMR (400 MHz, CDCl3): δ = 7.63–7.59 (m, 2 H), 7.40–7.37 (m,
4 H), 7.29–7.16 (m, 8 H), 4.11 (d, 1JH,P = 202.5 Hz, 2 H), 2.21 (q,
3JH,H = 7.3 Hz, 4 H), 1.76 (s, 6 H), 1.31 (s, 6 H), 0.90 (t, 3JH,H = 7.3
Hz, 6 H).
13C{1H} NMR (100 MHz, CDCl3): δ = 153.0, 150.3 (br), 139.9,
136.8, 135.1, 134.9 (d, JC,P = 15.3 Hz), 133.8, 132.8, 130.6, 129.0,
128.8 (d, JC,P = 5.6 Hz), 127.1, 125.4, 17.3, 14.7, 14.5, 12.1.
13C{1H} NMR (100 MHz, CDCl3): δ = 153.4, 150.1 (br), 141.3,
139.0, 134.8, 133.8, 133.1, 132.1 (d, JC,P = 10.4 Hz), 130.2, 129.1
(d, JC,P = 15.2 Hz), 128.8 (d, 1JC,P = 187.9 Hz), 127.2, 125.5, 62.2 (d,
2JC,P = 5.7 Hz), 17.3, 16.3 (d, 3JC,P = 5.8 Hz), 14.6, 14.7, 12.1.
31P{1H} NMR (162 MHz, CDCl3): δ = 18.7.
11B{1H} NMR (128 MHz, CDCl3): δ = –1.1.
31P{1H} NMR (162 MHz, CDCl3): δ = –121.5 (tt, 1JP,H = 202.5 Hz,
HRMS (NSI+): m/z [M + H]+ calcd for C39H47BN2O3P: 632.3448;
found: 632.3447.
3JP,H = 7.6 Hz).
11B{1H} NMR (128 MHz, CDCl3): δ = –1.0.
HRMS (APCI+): m/z [M + H]+ calcd for C35H39BN2P: 528.2975;
UV (THF): λmax (ε) = 518 nm (83 000).
8-[(4-Diethylphosphonato)phenyl]-4,4-dimethyl-1,3,5,7-te-
tramethyl-2,6-diethyl-4-bora-3a,4a-diaza-s-indacene (5b)
Phosphonate 6 (7.3 g, 14.1 mmol) was dissolved in anhydrous THF
(200 mL). To this solution was added MeMgBr (9.9 mL, 29.7
mmol, 3 M solution in Et2O) dropwise at r.t. The solution was
stirred at r.t. until complete consumption of starting material was
observed by TLC (3 h). The reaction was quenched with MeOH (30
mL) and the solvent was evaporated to leave a red solid. Purification
was performed by column chromatography on silica gel (EtOAc–
PE, 3:1) to give the intended product; yield: 6.18 g (86%); orange
solid; mp 91–95 °C, Rf = 0.4 (EtOAc–PE, 3:1).
found: 528.2970.
UV (THF): λmax (ε) = 518 nm (79 000).
8-[(4-Phosphino)phenyl]-4,4-dimethyl-1,3,5,7-tetramethyl-2,6-
diethyl-4-bora-3a,4a-diaza-s-indacene (1b)
Purification was performed by column chromatography on silica
gel (CHCl3–hexanes, 1:4) to afford the intended product, yield: 6.69
g (84%); orange solid; mp 196–198 °C; Rf = 0.4 (CHCl3–hexanes,
1:4).
IR (neat): 2958 (w), 2925 (w), 2361 (w, P–H), 2341 (w, P–H), 1551
(s), 1470 (s), 1531 (m), 1167(s), 1143 (m), 1110 (m), 1060 (m), 942
cm–1 (s).
1H NMR (500 MHz, CDCl3): δ = 7.60 (m, 2 H), 7.26 (m, 2 H), 4.11
(d, 1JH,P = 202.5 Hz, 2 H), 2.46 (s, 6 H), 2.32 (q, 3JH,H = 7.6 Hz, 4
H), 1.28 (s, 6 H), 0.99 (t, 3JH,H = 7.6 Hz, 6 H), 0.29 (s, 6 H).
13C{1H} NMR (100 MHz, CDCl3): δ = 150.8, 139.9, 137.3, 134.9
(d, JC,P = 15.4 Hz), 133.8, 132.6, 129.0, 128.9 (d, JC,P = 5.9 Hz),
128.8, 17.6, 14.8, 14.4, 12.1, 10.5 (br).
IR (neat): 2960 (w), 2931 (w), 1556 (s), 1453 (m), 1360 (m), 1322
(m), 1244 (m), 1172 (m), 1047 (m), 1014 (w), 945 cm–1 (s).
3
1H NMR (400 MHz, CDCl3): δ = 7.89 (dd, JH,H = 7.8 Hz, 3JH,P
=
13.1 Hz, 2 H), 7.42 (dd, 3JH,H = 7.8 Hz, 4JH,P = 3.8 Hz, 2 H), 4.19–
3
4.05 (m, 4 H), 2.40 (s, 6 H), 2.25 (q, JH,H = 7.3 Hz, 4 H), 1.29 (t,
3JH,H = 7.0 Hz, 6 H), 1.19 (s, 6 H), 0.92 (t, 3JH,H = 7.4 Hz, 6 H), 0.24
(s, 6 H).
13C{1H} NMR (100 MHz, CDCl3): δ = 150.9, 141.5 (d, JC,P = 2.7
Hz), 138.7, 133.3, 132.6, 132.0 (d, JC,P = 10.0 Hz), 129.0 (d, JC,P
=
31P{1H} NMR (202 MHz, CDCl3): δ = –121.7 (tt, 1JP,H = 202.5 Hz,
15.3 Hz), 128.5 (d, 1JC,P = 188.6 Hz), 128.4, 62.0 (d, 2JC,P = 5.7 Hz),
3JP,H = 7.4 Hz).
17.3, 16.2 (d, 3JC,P = 5.8 Hz), 14.6, 14.2, 11.8, 10.3 (br).
11B{1H} NMR (128 MHz, CDCl3): δ = –2.1.
HRMS (NSI+): m/z [M + H]+ calcd for C25H35BN2P: 404.2662;
31P{1H} NMR (162 MHz, CDCl3): δ = 18.7.
11B{1H} NMR (128 MHz, CDCl3): δ = –1.7.
found: 404.2665.
HRMS (NSI+): m/z [M + H]+ calcd for C29H43BN2O3P: 508.3135;
UV (THF): λmax (ε) = 512 nm (79 000).
found: 508.3129.
UV (THF): λmax (ε) = 513 nm (91 000).
Acknowledgment
Reduction of Phosphonates: General Procedure
We thank the EPSRC for a Career Acceleration Fellowship (L.J.H.),
a KTA award with High Force Research Ltd (J.F.W.), and its Natio-
nal Mass Spectrometry Service Centre, Swansea, UK and Newcast-
le University for a Studentship (L.H.D.). We also thank Johnson
Matthey for the loan of palladium acetate.
A THF solution of LiAlH4 (49.2 mL, 49.2 mmol, 1 M solution in
THF) was cooled to –78 °C. Me3SiCl (6.2 mL, 49.2 mmol) was add-
ed and the reaction mixture was warmed up to r.t. over 30 min. The
solution was cooled to –78 °C and the appropriate phosphonate 5
(19.7 mmol) in THF (1000 mL) was added slowly to the reaction
mixture. The solution was allowed to warm up to r.t. and stirred for
3 h. The mixture was then concentrated in vacuo and degassed H2O
(50 mL) was added dropwise to quench the reaction after first cool-
Supporting Information for this article is available online
at
http://www.thieme-connect.com/products/ejournals/journal/
10.1055/s-00000084.SunogIpimrfiantoSuIpg
n
fonirtat
ori
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 2622–2628

2628
L. H. Davies et al.
PAPER
(13) (a) Mésangeau, C.; Yous, S.; Pérès, B.; Lesieur, D.; Besson,
T. Tetrahedron Lett. 2005, 46, 2465. (b) Ficks, A.; Sibbald,
C.; Ojo, S.; Harrington, R. W.; Clegg, W.; Higham, L. J.
Synthesis 2013, 45, 265.
(14) (a) Shen, Z.; Röhr, H.; Rurack, K.; Uno, H.; Spieles, M.;
Schulz, B.; Reck, G.; Ono, N. Chem. Eur. J. 2004, 10, 4853.
(b) Ziessel, R.; Goze, C.; Ulrich, G.; Césario, M.; Retailleau,
P.; Harriman, A.; Rostron, J. P. Chem. Eur. J. 2005, 11,
7366.
(15) The quantum yield was referenced to 4,4-difluoro-8-phenyl-
1,3,5,7-tetramethyl-2,6-diethyl-4-bora-3a,4a-diaza-s-
indacene, (φF = 0.76 in THF (λabs = 524 nm, λem = 537 nm,
ε = 86 000 M–1cm–1) in: Gabe, Y.; Urano, Y.; Kikuchi, K.;
Kojima, H.; Nagano, T. J. Am. Chem. Soc. 2004, 126, 3357.
(16) (a) Jurisson, S. S.; Benedict, J. J.; Elder, R. C.; Deutsch, E.
Inorg. Chem. 1983, 22, 1332. (b) Torres Martin de Rosales,
R.; Finucane, C.; Mather, S. J.; Blower, P. J. Chem.
Commun. 2009, 4847. (c) Kowada, T.; Kikuta, J.; Kubo, A.;
Ishii, M.; Maeda, H.; Mizukami, S.; Kikuchi, K. J. Am.
Chem. Soc. 2011, 133, 17772. (d) Palma, E.; Correia, J. D.
G.; Oliveira, B. L.; Gano, L.; Santos, I. C.; Santos, I. Dalton
Trans. 2011, 40, 2787.
References
(1) (a) Maier, L. In Organic Phosphorus Compounds; Vol. 1;
Kosolapoff, G. M.; Maier, L., Eds.; Wiley-Interscience:
New York, 1972, 4. (b) Higham, L. J. Phosphorus
Compounds: Advanced Tools in Catalysis and Material
Sciences, In Catalysis by Metal Complexes; Vol. 37;
Peruzzini, M.; Gonsalvi, L., Eds.; Springer: Berlin, 2011, 1.
(2) (a) Brynda, M. Coord. Chem. Rev. 2005, 249, 2013.
(b) Katti, K. V.; Pillarsetty, N.; Raghuraman, K. Top. Curr.
Chem. 2003, 229, 121.
(3) (a) Yoshifuji, M.; Shibayama, K.; Inamoto, N.; Matsushita,
T.; Nishimoto, K. J. Am. Chem. Soc. 1983, 105, 2495.
(b) Yoshifuji, M.; Shibayama, K.; Toyota, K.; Inamoto, N.
Tetrahedron Lett. 1983, 24, 4227.
(4) (a) Henderson, W.; Alley, S. R. J. Organomet. Chem. 2002,
656, 120. (b) Pillarsetty, N.; Raghuraman, K.; Barnes, C. L.;
Katti, K. V. J. Am. Chem. Soc. 2005, 127, 331.
(5) (a) Davies, L. H.; Stewart, B.; Harrington, R. W.; Clegg, W.;
Higham, L. J. Angew. Chem. Int. Ed. 2012, 51, 4921.
(b) Note the presence of peroxides will oxidize the
phosphines and thus these should be avoided (aged ether
solvents, sunlight, etc.).
(17) Kyba, E. P.; Liu, S. T.; Harris, R. L. Organometallics 1983,
2, 1877.
(6) Stewart, B.; Harriman, A.; Higham, L. J. Organometallics
2011, 30, 5338.
(18) Note: direct reduction of phosphonate 6 even in dilute
solution caused decomposition of the Bodipy core.
(19) (a) Hiney, R. M.; Higham, L. J.; Müller-Bunz, H.; Gilheany,
D. G. Angew. Chem. Int. Ed. 2006, 45, 7248. (b) Bartlett, R.
A. M.; Olmstead, M.; Power, P. P.; Sigel, G. A. Inorg. Chem.
1987, 26, 1941. (c) Reiter, S. A.; Assmann, B.; Nogai, S. D.;
Mitzel, N. W.; Schmidbaur, H. Helv. Chim. Acta 2002, 85,
1140.
(20) CrysAlisPro; Agilent Technologies Ltd: Oxford (UK), 2013.
(21) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J.
A. K.; Puschmann, H. J. Appl. Cryst. 2009, 42, 339.
(22) Sheldrick, G. M. Acta Crystallogr. Sect., A 2008, 64, 112.
(23) The supplementary crystallographic data CCDC 967537–
967539 can be obtained free of charge at
(7) Loudet, A.; Burgess, K. Chem. Rev. 2007, 107, 4891.
(8) Benniston, A. C.; Copley, G.; Elliott, K. J.; Harrington, R.
W.; Clegg, W. Eur. J. Org. Chem. 2008, 16, 2705.
(9) Please note that if lower equivalents are used the yield is
significantly reduced, e.g., i-Pr2NEt (3 equiv) and Et2O·BF3
(4 equiv) yields only 10% of 3.
(10) For comprehensive reviews, see: (a) Tappe, F. M. J.;
Trepohl, V. T.; Oestreich, M. Synthesis 2010, 3037.
(b) Beletskaya, I. P.; Kazankova, M. A. Russ. J. Org. Chem.
2002, 38, 1391. (c) Van der Jeught, S.; Stevens, C. V. Chem.
Rev. 2009, 109, 2672.
(11) (a) Kalek, M.; Stawinski, J. Organometallics 2007, 26,
5840. (b) Kalek, M.; Stawinski, J. Organometallics 2008,
27, 5876. (c) Kohler, M. C.; Grimes, T. V.; Wang, X.;
Cundari, T. R.; Stockland, R. A. Jr. Organometallics 2009,
28, 1193.
www.ccdc.cam.ac.uk/conts/retrieving.html or from the
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge, CB2 1EZ, UK; fax: +44(1223)336033; E-mail:
>deposit@ccdc.cam.ac.uk.
(12) Kalek, M.; Ziadi, A.; Stawinski, J. Org. Lett. 2008, 10, 4637.
Synthesis 2014, 46, 2622–2628
© Georg Thieme Verlag Stuttgart · New York