Reaction of 1,3-diynes with the electron rich iron complex (η5-C5Me5)(η2-dppe)Fe-Cl. A new and direct access to the iron butadiynyl complex (η5-C5Me5)(η 2-dppe)Fe-CCCCSiMe3

Article abstract of DOI:10.1016/S0022-328X(01)00797-5

The electron rich iron complexes (C₅Me₅)Fe(dppe)Cl (3) and (C₅Me₅)Fe(dppe)Br (4) do not react with Li(CC)_nSiMe₃ (n=1, 2). Treatment of the chloro iron complex 3 with Me₃SiCC-CCSiMe₃ in methanol and in the presence of NaBPh₄ provides the binuclear complex (C₅Me₅)(dppe)FeCC-CC-Fe(dppe)(C₅Me₅) (10) as the unique product of this clean reaction (73%). Reaction of the chloro iron derivative 3 with one equivalent of trimethylsilyl(1,3-butadiyne) in triethylamine as a solvent and in the presence of NaBPh₄ yields 82% of the mononuclear iron trimethylsilyl-1,3-butadiyne compound. The triethylamine acts as a base to selectively deprotonate the putative butatrienylidene intermediate.

Full text of DOI:10.1016/S0022-328X(01)00797-5

Journal of Organometallic Chemistry 629 (2001) 39–43
www.elsevier.com/locate/jorganchem
Reaction of 1,3-diynes with the electron rich iron complex
(h5-C5Me5)(h2-dppe)FeꢀCl. A new and direct access to the iron
butadiynyl complex (h5-C5Me5)(h2-dppe)FeꢀCꢁCCꢁCSiMe3
Franc¸oise Coat, Patrice Thominot, Claude Lapinte *
Department of Chemistry, Organome´talliques et Catalyse: Chimie et Electrochimie Mole´culaire, UMR CNRS 6509,
Institut de Chimie de Rennes, Uni6ersite´ de Rennes 1, Campus de Beaulieu, 35042 Rennes Cedex, France
Received 8 February 2001; accepted 26 March 2001
Abstract
The electron rich iron complexes (C5Me5)Fe(dppe)Cl (3) and (C5Me5)Fe(dppe)Br (4) do not react with Li(CꢁC)nSiMe3 (n=1,
2). Treatment of the chloro iron complex 3 with Me3SiCꢁCꢀCꢁCSiMe3 in methanol and in the presence of NaBPh4 provides the
binuclear complex (C5Me5)(dppe)FeCꢁCꢀCꢁCꢀFe(dppe)(C5Me5) (10) as the unique product of this clean reaction (73%). Reaction
of the chloro iron derivative 3 with one equivalent of trimethylsilyl(1,3-butadiyne) in triethylamine as a solvent and in the presence
of NaBPh4 yields 82% of the mononuclear iron trimethylsilyl-1,3-butadiyne compound. The triethylamine acts as a base to
selectively deprotonate the putative butatrienylidene intermediate. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: (C5Me5)(dppe)FeCꢁCꢀCꢁCꢀFe(dppe)(C5Me5); (C5Me5)Fe(dppe)(CꢁCꢀCꢁCSiMe3); Triethylamine
1. Introduction
allowing the preparation of a variety of transition metal
compounds containing the trimethylsilyl 1,3-butadiyne
ligand [1]. In the iron series, conversion of
(C5R5)Fe(CO)2X (R=H, Me, Ph) to the corresponding
trimethylsilylbutadiynyl complexes has been achieved
following this route (Scheme 1) [5–8]. On the other
hand, terminal 1,3-diynes can oxidatively add to the
electronically unsaturated metal centers, for instance,
RhCl(PPri3)2 and HCꢁCCꢁCR gave the Rh(III)–diynyl
complex RhHCl(CꢁCCꢁCR)(PPri3)2 [9,10]. Moreover,
in the presence of NEt3, HCꢁCCꢁCR reacts with cis-
RuCl2(dppe)2 to give trans-RuCl(CꢁCCꢁCR)(dppe)2
[11,12].
In connection with our ongoing interest in the syn-
thesis and physical properties of molecular wires in
which an all carbon chain is capped by two redox active
metal termini, we need to find out reactions to prepare
the mononuclear precursors containing polyynes lig-
ands [13]. A few years ago, a three-step access to the
binuclear complex (C5Me5)(dppe)FeCꢁCꢀCꢁCꢀFe-
(dppe)(C5Me5) (10) [14] and a two-step route to
(C5Me5)(dppe)Fe(CꢁCꢀCꢁCSiMe3) (11) [7] were re-
ported by our group. We now report the reactions of
the electron rich (C5Me5)(dppe)FeCl complex with
Despite the wide interest of chemists pertaining to the
reactions of alkynes with transition metal complexes,
the expansion of this branch chemistry to diynes and
polyynes is still in its infancy [1]. The products formed
by the reaction of 1,3-diynes with transition metal
complexes appear strongly dependent on the nature of
the metal, its electron richness and steric environment.
The lithium salt of the trimethylsilyl-protected 1,3-bu-
tadiyne, Me3SiCꢁCꢀCꢁCLi·LiBr, readily obtained by
treating the bistrimethylsilyl analogue Me3SiCꢁCꢀ
CꢁCSiMe3 with MeLi·LiBr [2,3] constitutes a very use-
ful synthon to prepare [M]CꢁCꢀCꢁCSiMe3 derivatives.
As the relative nucleophilicity of Li(CꢁC)nSiMe3 de-
creases with the carbon chain length, this reagent is less
nucleophilic than LiCꢁCSiMe3. For instance, in con-
trast with this latter, it does not react with M(CO)6
(M=Cr, W) [4]. However, it is nucleophilic enough to
displace halide from many transition metal complexes
* Corresponding author. Tel.: +33-99-285963; fax: +33-99-
285963.
E-mail address: claude.lapinte@univ-rennesl.fr (C. Lapinte).
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(01)00797-5

40
F. Coat et al. / Journal of Organometallic Chemistry 629 (2001) 39–43
of 1-alkynes with a wide variety of transition metal
halides via a 1,2-H shift [17,18]. This reaction was
extended to terminal 1,3-diynes to prepare transition
metal complexes containing a 1,3-dyine ligand [11,12].
However, this extension is limited to a small number of
complexes with an electron poor metal center. In the
cases of more electron-rich fragments, 1,3-diynes tend
to undergo a 1,4-H shift affording highly reactive buta-
trienylidene transition metal intermediates, which are
not stable in the reaction conditions [1,19–22]. Thus,
treatment of 3 with Me3SiCꢁCH in methanol and in the
presence of the potassium hexafluorophosphate salt,
Me3SiCꢁCꢀCꢁCLi·LiBr,
Me3SiCꢁCꢀCꢁSiMe3
and
Me3SiCꢁCꢀCꢁCH in different experimental conditions.
These investigations allowed us to find a one-step syn-
thesis of the key compounds (10) and (11).
2. Results and discussion
In contrast with (C5Me5)Fe(CO)2Br [7], (C5Me5)-
Fe(dppe)Cl (3) or (C5Me5)Fe(dppe)Br (4) do not react
with Li(CꢁC)xSiMe3. Indeed, when this reaction is car-
ried out at low temperature in THF no reaction oc-
curred. Nevertheless, at 20°C in THF, a color change
from dark green to brown resulted from the treatment
of 3 with Me3SiCꢁCꢀCꢁCLi·LiBr. After 4 h of reaction,
three organoiron complexes are present in the reaction
mixture: the halide derivatives 3 and 4 in the ratio 60:40
and a small amount (5%) of the known iron hydride
(C5R5)Fe(dppe)H (5) (Scheme 1) [15]. It is probable
that the inertness of 3 toward Li(CꢁC)2SiMe3 could be
due to the weak electrophilic character of the iron
center, but also to a strong steric protection around the
metal provided by both the C5Me5 and dppe ligands
[16]. So, alternative routes must be found to
(C5Me5)Fe(dppe)CꢁCꢀCꢁCSiMe3.
produces
vinylidene
[(C5Me5)Fe(dppe)(=C=
CH2)][PF6] (6) in 95% yield [14] (Scheme 2), but the
replacement of Me3SiCꢁCH by Me3SiCꢁCꢀCꢁCH af-
fords, in the same conditions, the allenylidene iron
complex [(C5Me5)Fe(dppe)(=C=C=C(CH3)OMe)]-
[PF6] (7) via butatrienylidene [(C5Me5)Fe(dppe)(=C=
C=C=C(H)SiMe3)]X (8) [21]. Similar to the reaction
observed in the case of the vinylidene, the cleavage of
the trimethylsilyl fragment should occur after the hy-
drogen 1,4-shift and a subsequent addition of a
methanol molecule provides the allenylidene 7 (Scheme
2) [19].
We also found that treatment of Me3SiCꢁCꢀC6H5
with one equivalent of 3 in the methanol–THF (10:1)
mixture and in the presence of KF and KPF6 (NaBPh4
can also be used) salts produces the iron alkynyl deriva-
tive (C5Me5)Fe(dppe)(CꢁCꢀC6H5) (9) isolated in 80%
yield (Scheme 3) [23]. In this one-step procedure the
The deprotonation of vinylidene complexes consti-
tutes one of the most general methods to prepare
transition metal compounds featuring acetylide ligands,
themselves being conveniently obtained from reactions
Scheme 1.
Scheme 2.

F. Coat et al. / Journal of Organometallic Chemistry 629 (2001) 39–43
41
Scheme 5.
Scheme 3.
ethynyl benzene is in situ generated by the fluoride-in-
duced cleavage of the terminal trimethylsilyl group. As
extensively described, the PF6 anion promotes the com-
plexation of the terminal alkyne at the iron center
giving the vinylidene derivative isolated in most cases as
the final product of the reaction. However, in the
above-defined conditions, the cleavage of the
trimethylsilyl group is associated with the formation of
a stoichiometric amount of a strong base namely
MeOK. This base, in situ deprotonates the vinylidene
intermediate, allowing the formation of 9 as the unique
organoiron product of this sequence of reactions. By
analogy, we though that treatment of the chloro iron
complex 3 with Me3SiCꢁCꢀCꢁCSiMe3 in the same con-
ditions, could give the 1,3-butadiynyl complex
(C5Me5)Fe(dppe)(CꢁCꢀCꢁCSiMe3) (11). However, the
mononuclear compound was not formed in these condi-
tions. The known binuclear complex (C5Me5)(dppe)-
FeCꢁCꢀCꢀCꢀFe(dppe)(C5Me5) (10) constitutes the
unique product of this clean reaction (Scheme 4). Inter-
estingly, compound 10 is isolated in 73% yield. This
reaction constitutes a new synthetic access to the binu-
clear complex 10 much more directly than the oxidative
of a small amount of the hydride 5 beside the bis(iron)
derivative 10 (Scheme 4).
Assuming that the formation of the iron allenylidene
7 results from the fast addition of a nucleophilic
molecule of solvent onto the butatrienylidene 6 (Scheme
2), we thought that the use of a more basic and less
nucleophilic solvent could avoid the nucleophilic addi-
tion. We and the others found that primary and sec-
ondary amines react with in situ generated
butatrienyllidene complexes providing amino-allenyli-
denes [19,20,24]. However, the use of bulky and weakly
nucleophilic amine should favor the deprotonation of
the transient butatrienylidene intermediate with respect
to the nucleophilic addition of the amine. Accordingly,
we reacted the chloro iron compound 3 with one equiv-
alent of trimethylsylil(1,3-butadiyne) in diethylamine as
a solvent. Analysis of the products after the reaction
indicates the presence of a neutral compound and sev-
eral organoiron iron salts, which were not identified.
The neutral compound was isolated in 33% yield from
a CH2Cl2 liquor containing the reaction products after
the precipitation of the salts by the addition of diethyl
ether. It was identified as the desired derivative
(C5Me5)(dppe)Fe(CꢁCꢀCꢁCSiMe3) (11). As the forma-
tion of salts should result from an initial nucleophilic
addition of the secondary amine on the putative buta-
trienylidene intermediate, the reaction was reproduced
in the same conditions, but using triethylamine as a
solvent. In this case, the solvent selectively acts as a
base and deprotonates the butatrienylidene intermedi-
ate, providing the target complex 11 in 82% yield
(Scheme 5).
coupling
of
the
iron
alkynyl
derivative
(C5Me5)Fe(dppe)(CꢁCH) [14]. It is noteworthy, that the
use of two or three equivalents of Me3SiCꢁCꢀCꢁ
CSiMe3 did not allow us to observe the presence of any
mononuclear species. Assuming that the complexation
of the second iron center could occur on the buta-
trienylidene intermediate, we tried to perform the reac-
tion in the presence of one equivalent of potassium
tert-butoxide and two equivalents of Me3SiCꢁCꢀCꢁ
CSiMe3 to favor both the formation of a mononuclear
intermediate and its quick deprotonation. However, we
did not observe the formation of (C5Me5)(dppe)FeCꢁ
CꢀCꢁCSiMe3. The unique difference was the formation
In conclusion, we found direct routes to
(C5Me5)(dppe)FeCꢁCꢀCꢁCꢀFe(dppe)(C5Me5) (10) and
(C5Me5)Fe(dppe)(CꢁCꢀCꢁCSiMe3) (11) which are now
easily accessible in fair yield. Our results show that
Scheme 4.

42
F. Coat et al. / Journal of Organometallic Chemistry 629 (2001) 39–43
terminal alkynes and diynes present a similar reactivity.
The difference of behavior between these two families
of compounds comes mainly from the high reactivity of
butatrienylidene intermediates in comparison with the
relative chemical inertness of the vinylidene derivatives.
For this reason, the complexation of diynes at an
electron rich metal center require accurately defined
reaction conditions. Following the strategy reported
here, we think that it should be possible to prepare
butadiynyl complexes with different metal centers and a
wide variety of 1,3-diynes [25].
C5Me5)Fe(dppe)Br (4), and (h-C5Me5)Fe(dppe)H (5),
in a 38:57:5 ratio. Complexes 3, 4, and 5 were identified
by comparison of the 1H-NMR spectra and cyclic
voltammograms with those of authentic samples [26].
3.3. (p5-C5Me5)(p5-dppe)FeꢀCꢁCꢀCꢁCꢀFe(p5-dppe)-
(p5-C5Me5) (10)
3.3.1. In a Schlenk tube the complex 3 (0.250 g, 0.40
mmol) and 20 ml of methanol were introduced. The
suspension was cooled down to −25°C before adding
0.6 equivalents of trimethylsylyl-1,3-butadiyne (0.030 g,
0.24 mmol), 1.1 equivalents of NaBPh4 (0.150 g, 0.44
mmol), and 1.1 equivalents of KOBut (0.0493 mg, 0.44
mmol), successively. The mixture was allowed to warm
up to 20°C in 16 h upon stirring. Initially dark green,
the solution turned orange. Evaporation of the solvent
under vacuum and extraction of the solid residue with
a toluene–diethyl ether 1:1 mixture provided an orange
3. Experimental
3.1. General data
All manipulations were carried out under argon at-
mosphere. Solvents or reagents were used as follows:
Et2O and n-pentane, distilled from Na/benzophenone;
CH2Cl2, distilled from CaH2 and purged with argon;
HN(iPr)2, and NEt3, distilled from KOH and purged
with argon. Complex 3 [26] and trimethylsilyl-1,3-bu-
tadiyne [27] were prepared following reported proce-
dures. High field NMR spectra experiments were
performed on multinuclear Bruker 300 or 200 MHz
instruments (AM300WB and 200DPX). Chemical shifts
are given in parts per million relative to tetramethylsi-
1
solid residue. Comparison of the IR, H, 31P and CV
data of the crude products with those of authentic
samples indicated the formation of the two iron com-
plexes (h-C5Me5)Fe(dppe)H (5) [26] and (C5Me5)-
(dppe)FeCꢁCꢀCꢁCꢀFe(dppe)(C5Me5) (10) [14] in the
60:40 ratio as determined from the integration of the
1H-NMR spectra.
3.3.2. In a Schlenk tube 1,4-bis(trimethylsilyl-1,3-bu-
tadiyne (0.055 g, 0.28 mmol) and 1 ml of THF were
introduced. To this solution, complex 3 (0.350 g, 0.56
mmol), methanol (20 ml), KF (0.033 mg 0.56 mmol)
and Na BPh4 (0.191 g, 0.56 mmol) were successively
introduced. The resulting suspension was stirred under
reflux for 4 h. A slow color change from dark green to
brown was observed. Evaporation of the solvent under
vacuum and extraction of the crude residue with 4×5
ml of toluene provided a brown powder. After washing
with 5 ml of diethyl ether the solid was identified as the
pure binuclear complex (10) isolated in 73% yield.
1
lane (TMS) for H- and 13C-NMR spectra, H3PO4 for
31P-NMR spectra. Transmittance-FTIR spectra were
recorded using a Bruker IFS28 spectrometer (400–4000
cm−1). UV–vis spectra were recorded on an UVIKON
942 spectrometer. Cyclic voltammograms were recorded
using a PAR 263 in CH2Cl2 (0.1 M (n-Bu)4N+ PF6−) at
25°C at a platinum electrode, using a SCE reference
electrode and ferrocene as internal calibrant (0.460 V)
[28].
3.2. Reaction of TMSCꢁCCꢁCLi·LiBr with
(p5-C5Me5)(p5-dppe)FeCl
3.4. (p5-C5Me5)(p2-dppe)FeꢀCꢁCꢀCꢁCꢀSiMe3 (11)
In a Schlenk tube bis(trimethylsilyl)-1,3-butadiyne
(0.117 g, 0.60 mmol) and THF (15 ml) were introduced.
The resulting solution was cooled down to −80°C
before adding MeLi·LiBr (1.65 ml, 0.75 mmol). The
solution was then allowed to warm up to 20°C and
stirred for 4 h. Separately, in another Schlenk tube, the
green complex 3 (0.356 mg, 0.56 mmol), KPF6 (0.112 g,
0.6 mmol) and 30 ml of methanol were introduced. This
suspension was cooled down to −80°C and the solu-
tion of THF containing the trimethylsilyl-1,3-bu-
tadiynediyl lithium was transferred. The solution was
allowed to warm up upon stirring for 16 h. Evaporation
of the solvent under vacuum and extraction of the solid
residue with CH2Cl2 provided a greenish brown powder
containing the complexes (h-C5Me5)Fe(dppe)Cl (3), (h-
3.4.1. To a suspension of 3 [26] (0.300 g, 0.48 mmol)
and NaBPh4 (0.180 g, 0.58 mmol) in diethylamine (25
ml), 1.2 equivalents of trimethylsilylbutadiyne (0.070 g,
0.55 mmol) was added. The mixture is stirred at 20°C
for 16 h. After evaporation of the solvent under re-
duced pressure, the crude residue was extracted with
4×5 ml CH2Cl2 and concentrated to 10 ml. After the
precipitation of non-identified organoiron salts by the
addition of diethyl ether and filtration, evaporation to
dryness of the filtrate provide an orange powder iden-
tified by comparison with the data of an authentic
sample as the pure compound 11 (33%) [7].
3.4.2. To a suspension of 3 [26] (0.300 g, 0.48 mmol)
and NaBPh4 (0.185 g, 0.54 mmol) in triethylamine (25
ml), 1.2 equivalents of trimethylsilylbutadiyne (0.070 g,

F. Coat et al. / Journal of Organometallic Chemistry 629 (2001) 39–43
43
[12] C. Lebreton, D. Touchard, L. Le Pichon, A. Daridor, L. Toupet,
P.H. Dixneuf, Inorg. Chim. Acta 272 (1998) 188.
[13] F. Paul, C. Lapinte, Coord. Chem. Rev. 178–180 (1998) 427.
[14] N. Le Narvor, L. Toupet, C. Lapinte, J. Am. Chem. Soc. 117
(1995) 7129.
[15] The formation of the iron hydride 5 should result from the
reduction of 3 by the organolithium reagent which should give
rise to the 17-electron species {(C5Me5)(dppe)Fe} by loss of the
chloride anion. It has been shown that this 17-electron complex
can abstract a hydrogen atom from the solvent [29].
0.55 mmol) was added. While the mixture was stirred at
20°C for 16 h, the dark green color slowly turned to
orange. After removal of the solvent, the solid residue
was extracted with 30 ml (3×10 ml) of diethyl ether.
The solvent was removed under reduced pressure and
the resulting orange powder was dried under vacuum.
The product was identified as a pure sample of 11
(82%) by comparison with the spectroscopic data of an
authentic sample [7].
[16] When the dppe ligand is replaced by the less bulky dippe, the
reaction
of
Li(CꢁC)2SiMe3
with
3
produces
(C5Me5)Fe(dippe)(CꢁCꢀCꢁCꢀSiMe3) in fair yield [30].
[17] M.I. Bruce, A.G. Swincer, Adv. Organomet. Chem. 52 (1987)
3940.
Acknowledgements
[18] M.I. Bruce, Chem. Rev. 91 (1991) 197.
[19] M.I. Bruce, P. Hinterding, P.J. Low, B.W. Skelton, A.H. White,
Chem. Commun. (1996) 1009.
[20] M.I. Bruce, P. Hinterding, M. Ke, P.J. Low, B.W. Skelton, A.H.
White, Chem. Commun. (1997) 715.
We thank the French Ministe`re MENRT and la
Re´gion Bretagne for award of thesis grants to F.C. and
P.T., respectively.
[21] (a) V. Guillaume, P. Thominot, F. Coat, A. Mari, C. Lapinte, J.
Organomet. Chem. 565 (1998) 75;
References
(b) F. Coat, M. Guillemot, F. Paul, C. Lapinte, J. Organomet.
Chem. 578 (1999) 76.
[1] P.J. Low, M.I. Bruce, Adv. Organomet. Chem. 48 (2001) 000.
[2] A.B. Holmes, C.L.D. Jennings-White, A.H. Schulthess, B.
Akinde, D.R.M. Walton, J. Chem. Soc. Chem. Commun. (1979)
840.
[3] J.L. Brefort, R.J.P. Corriu, P. Gerbier, C. Guerin, B. Henner, A.
Jean, T. Kuhlmann, F. Garnier, A. Yassar, Organometallics 11
(1992) 2500.
[4] C. Hartbaum, H. Fischer, Chem. Ber./Recueil 130 (1997) 1063.
[5] A. Wong, P.C.W. Kang, C.D. Tagge, D.R. Leon, Organometal-
lics 9 (1990) 1992.
[6] F. Coat, C. Lapinte, Organometallics 15 (1996) 477.
[7] F. Coat, M.-A. Guillevic, L. Toupet, F. Paul, C. Lapinte,
Organometallics 16 (1997) 5988.
[8] M. Akita, M.-C. Chung, A. Sakurai, S. Sugimoto, M. Terada,
M. Tanaka, Y. Moro-oka, Organometallics 16 (1997) 4882.
[9] H. Werner, O. Gevert, P. Steinert, J. Wolf, Organometallics 14
(1995) 1786.
[10] R.W. Lass, P. Steinert, J. Wolf, H. Werner, Chem. Eur. J. 2
(1996) 19.
[22] A remarkably stable butatrienylidene iridium complex was re-
cently reported: K. Ilg, H. Werner, Angew. Chem. Int. Ed. 39
(2000) 1632.
[23] (a) T. Weyland, C. Lapinte, G. Frapper, M.J. Calhorda, J.-F.
Halet, L. Toupet, Organometallics 16 (1997) 2024;
(b) J.R. Lomprey, J.P. Selegue, Organometallics 12 (1993) 616;
(c) M.I. Bruce, B.C. Hall, B.D. Kelley, P.J. Low, B.W. Skelton,
A.H. White, J. Chem. Soc. Dalton Trans. (1999) 3719.
[24] P. Tominot, C. Lapinte, unpublished results.
[25] This reaction was recently extended to the synthesis of a ruthe-
nium 1,3-butadiyne complex [31].
[26] C. Roger, P. Hamon, L. Toupet, H. Rabaaˆ, J.-Y. Saillard, J.-R.
Hamon, C. Lapinte, Organometallics 10 (1991) 1045.
[27] B. Bartik, R. Dembinski, T. Bartik, A.M. Arif, J.A. Gladysz,
New J. Chem. 21 (1997) 739.
[28] N.G. Connelly, W.E. Geiger, Chem. Rev. 96 (1996) 877.
[29] P. Hamon, L. Toupet, J.-R. Hamon, C. Lapinte, Organometal-
lics 15 (1996) 10.
[30] M. Guillemot, L. Toupet, C. Lapinte, Organometallics 17 (1998)
1928.
[11] P. Haquette, D. Touchard, L. Toupet, P.H. Dixneuf, J.
Organomet. Chem. 565 (1998) 63.
[31] M. Smith, M.I. Bruce, C. Lapinte, unpublished results.
.