Marked changes in relative nucleophilicity in comparing SN2AR reactions of free arenes and coordinated arenes. Kinetic studies of reactions of (η6-chlorobenzene) (η5-cyclopentadienyl)iron(II) tetrafluoroborate with anionic

Article abstract of DOI:10.1021/om9601817

Quantitative kinetic studies have been made of reactions of the (η⁶-chlorobenzene)(η⁵-cyclopentadienyl)iron(II) cation, (C₆H₅Cl)Fe(C₅H₅)⁺ (1), with methoxide, phenoxide, methanet

Full text of DOI:10.1021/om9601817

3198
Organometallics 1996, 15, 3198-3203
Ma r k ed Ch a n ges in Rela tive Nu cleop h ilicity in
Com p a r in g SN2Ar Rea ction s of F r ee Ar en es a n d
Coor d in a ted Ar en es. Kin etic Stu d ies of Rea ction s of
(η6-Ch lor oben zen e)(η5-cyclop en ta d ien yl)ir on (II)
Tetr a flu or obor a te w ith An ion ic a n d Neu tr a l
Nu cleop h iles
Paulo C. B. Gomes, Eduardo J . S. Vichi,* Paulo J . S. Moran,
Alberto Federman Neto,† Maria Lourdes Maroso, and J oseph Miller*,‡
Instituto de Quimica, Universidade Estadual de Campinas, CP 6154,
13083-970 Campinas SP, Brazil
Received March 7, 1996X
Quantitative kinetic studies have been made of reactions of the (η6-chlorobenzene)(η5-
cyclopentadienyl)iron(II) cation, (C6H5Cl)Fe(C5H5)+ (1), with methoxide, phenoxide, meth-
anethiolate, benzenethiolate, and azide ions in methanol, piperidine, morpholine, aniline,
and thiourea in methanol, and guanidine in ethanol. The results were compared with the
same nucleophile-solvent combinations in reactions with 1-chloro-2,4-dinitrobenzene,
C6H3(NO2)2Cl (2). In general, the reactivity decreases on passing from 2 to 1. The decrease
is small (e10-1) for phenoxide (PhO-), methanethiolate (MeS-), and guanidine (gua) and
large (ca. 10-5 ) for benzenethiolate (PhS-), azide (N3-), piperidine (pip), morpholine (morph),
aniline (anil), and thiourea (thiou). The differences in reactivity when comparing the SN-
2Ar reactions of 1 and 2 is discussed in terms of the different location of the negative charge
generated in the transition state by the electrons displaced from the reaction center by the
entering groups (arenide electrons).
In tr od u ction
The greater activating power of the [(η5-C5H5)Fe]+
4a
moiety5b compared with Cr(CO)3 includes a significant
Nucleophilic substitution reactions in halogenoarene
transition metal complexes have gained synthetic im-
portance in organic chemistry1 because the parent
organic molecules are virtually inert to nucleophiles
under normal conditions.2 The report by Nicholls and
Whiting3 that the Cr(CO)3 moiety of (η6-chlorobenzene)-
tricarbonylchromium activates methoxy dechlorination
led to the publication of many papers on SN2Ar reactions
activated by arene complexation to a transition metal.
These papers reported kinetic studies and synthetic
applications of halogenoarenes coordinated to transition
metal-ligand residues, such as Cr(CO)3,4 [(η5-C5H5)-
Fe]+,5 [(η5-C5H5)Ru]+,6 and [(CO)3Mn]+.7
It was established that π-coordination to Cr(CO)3
activates the halogenoarene ligand toward methoxide
to an extent similar in magnitude to the effect of a
4-nitro substituent,4a-c although different mechanisms
of electron withdrawal have been suggested for the two
systems.4e The effect of π-complexation with [(η5-
C5H5)Fe]+,5b [(CO)3Mn]+,7c and [(η5-C5H5)Ru]+ 6c is even
more marked; the effect of the [(η5-C5H5)Fe]+ moiety is
equivalent to that of o- and p-nitro groups combined.
entropic contribution, since the reaction of methoxide
with the iron(II) salt is between an anion and a cation.
In view of the interest in these SN2Ar reactions8 we
have studied the kinetics of the reactions of 1 with a
varied range of nucleophiles. In this paper we report
(4) (a) Brown, D. A.; Raju, J . R. J . Chem. Soc. A 1966, 40. (b)
Bunnett, J . F.; Herrmann, H. J . Org. Chem. 1971, 36, 4081. (c)
Tchissambon, L.; J ouen, G.; Dabard, R. C. R. Seances Acaad. Sci. 1972,
274C, 654, 806. (d) Semmelhack, M. F.; Hall, H. T. J . Am. Chem. Soc.
1974, 96, 7091, 7092. (e) Kreindlin, A. Z.; Khandarova, U. S.; Gubin,
S. P. J . Organomet. Chem. 1975, 92, 197. (f) Mahaffy, C. A. L.; Pauson,
P. L. J . Chem. Res., Synop. 1979, 128; J . Chem. Res., Miniprint 1979,
1773. (g) Semmelhack, M. F.; Clarck, G. R.; Garcia, J . L.; Harrison,
J .J .; Thebteraronth, Y.; Yamashita, A. Tetrahedron 1981, 37, 3959.
(h) Ostrowiski, S.; Mokoska, M. J . Organomet. Chem. 1989, 367, 95.
(5) (a) Nesmeyanov, A. N.; Vol’kenau, N. A.; Bolesova, I. N. Dokl.
Acad. Nauk SSSR 1967, 175, 106, 606. (b) Nesmeyanov, A. N.;
Vol’kenau, N. A.; Isaeva, L. S.; Bolesova, I. N. Dokl. Acad. Nauk SSSR
1968, 183, 184 and earlier papers cited therein. (c) Helling, J . F.;
Hendrickson, W. A. J . Organomet. Chem. 1979, 168, 87. (d) Khand, I.
V.; Pauson, P. L.; Watts, W. F. J . Chem. Soc. C 1969, 2024. (e) Lee, C.
C.; Azogu, C. I.; Chang, P. C.; Sutherland, R. G. J . Organomet. Chem.
1981, 220, 181. (f) Lee, C. C.; Gill, U. S.; Iqbal, M.; Azogu, C. I.;
Sutherland, R. G. J . Organomet. Chem. 1982, 231, 151. (g) Astruc, D.
Tetrahedron 1983, 39, 4027. (h) Moriarty, R. M.; Gill, U. S. Organo-
metallics 1986, 5, 253. (i) Lee, C. C.; Abd-el-Aziz, A. S.; Chowdhury,
R. L.; Gill, U. S.; Piorko, A.; Sutherland, R. G. J . Organomet. Chem.
1986, 315, 79. Sutherland, R. G.; Choudhury, R. L.; Piorko, A.; Lee, C.
C. Can. J . Chem. 1986, 64, 2031. (j) Terrier, F.; Vichard, D.; Cha-
trousse, A. P.; Top, S.; McGlinchey, M. J . Organometallics 1994, 13,
690.
(6) a) Sagall, J . A. J . Chem. Soc., Chem. Commun. 1985, 1338. (b)
Vol’kenau, N. A.; Bolesova, I. N.; Shulpina, L. S.; Kitaigorodskii, A.
N.; Kravtsov, D. N. J . Organomet. Chem. 1985, 288, 341. (c) Moriarty,
R. M.; Gill, U. S.; Ku, Y. Y. J . Organomet. Chem. 1988, 350, 157. (d)
Moriarty, R. M.; Ku, Y. Y.; Gill, U. S. Organometallics 1988, 7, 660.
(7) a) Walker, P. J . C.; Mawby, R. J . Inorg. Chem. 1971, 10, 404; J .
Chem. Soc., Dalton Trans. 1973, 622. (b) Pauson, P. L.; Segal, J . A. J .
Chem. Soc., Dalton Trans. 1975, 1677, 1683. (c) Knipe, A. C.; Guiness,
S. J .; Watts, W. E. J . Chem. Soc., Perkin Trans. 2 1981, 193. (d) Bhasin,
K. K.; Balkeen, W. G.; Pauson, P. L. J . Organomet. Chem. 1981, 204,
C25.
† Present address: Faculdade de Cieˆncias Farmaceuticas, Univer-
sidade de Sa˜o Paulo, 14040-900 Ribeira˜o Preto, SP, Brazil.
‡ Present address: Laborato´rio de Tecnologia Farmaceutica e De-
partamento de Qu´ımica/CCEN, Universidade Federal da Paraiba, CP
5009, 58051-970 J oa˜o Pessoa, Pb, Brazil.
X Abstract published in Advance ACS Abstracts, J une 1, 1996.
(1) See: (a) Semmelhack, M. F. J . Organomet. Chem. 1961, 1, 36.
(b) J ouen, G. In Transition Metal Organometallics in Organic Synthe-
sis; Alper, H., Ed.; Academic Press: New York, 1978; Vol. II, Chapter
2, and references therein.
(2) Miller, J . Aromatic Nucleophilic Substitution; Elsevier: Amster-
dam, London, and New York, 1968.
(3) Nicholls, B.; Whiting, M. C. J . Chem. Soc. 1956, 551.
S0276-7333(96)00181-1 CCC: $12.00 © 1996 American Chemical Society
+
+

SN2Ar Reactions of Free and Coordination Arenes
Organometallics, Vol. 15, No. 14, 1996 3199
Ta ble 1. Ra te Coefficien ts for Rea ction s of
(C6H5Cl)F e(C5H5)+ (1) a n d of 2,4-(NO2)2C6H3Cl (2)
w ith Nu cleop h iles
a
nucleophile
(solvent)
temp
(°C)
k2
substrate
(dm3 mol-1 s-1
)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
MeO- (MeOH)
PhO- (MeOH)
PhO- (MeOH)
PhO- (MeOH)
PhO- (MeOH)
PhO- (MeOH)
MeS- (MeOH)
MeS- (MeOH)
MeS- (MeOH)
PhS- (MeOH)
PhS- (MeOH)
PhS- (MeOH)
PhS- (MeOH)
PhS- (MeOH)
N3- (MeOH)
N3- (MeOH)
N3- (MeOH)
N3- (MeOH)
pipb (MeOH)
pip (MeOH)
13.6 2.41 × 10-2
20.8 1.72 × 10-4
25.2 2.95 × 10-4
31.2 6.03 × 10-4
33.2 7.23 × 10-4
41.4 1.83 × 10-3
10.9 4.08 × 10-3
quantitative and semiquantitative studies of reactions
with (a) azide (N3-), phenoxide (C6H5O-), methanethi-
olate (CH3S-), and benzenethiolate (C5H5S-) ions in
methanol, (b) piperidine (pip), morpholine (morph),
aniline (anil), and thiourea (thiou) in methanol, and (c)
guanidine (gua) in ethanol. The results are compared
with data for the samesor very similarsnucleophile-
solvent combinations in reactions with C6H3(NO2)2Cl
(2).9-16 Since there were no data in the literature for
the reactions of 2 with morpholine and guanidine, we
carried out a complete kinetic study of these reactions.
20.3 1.04 × 10-2, 1.07 × 10-2
30.5 2.74 × 10-2, 2.89 × 10-2
10.3 1.46 × 10-3
15.4 2.60 × 10-3, 2.73 × 10-3
20.4 4.36 × 10-3
25.0 7.62 × 10-3
29.9 1.30 × 10-2, 1.29 × 10-2
81.1 4.60 × 10-4, 4.74 × 10-4
85.2 7.29 × 10-4
89.0 1.22 × 10-3
90.0 1.27 × 10-3
102.9 3.54 × 10-3, 3.66 × 10-3
118.8 9.37 × 10-3, 9.43 × 10-3
138.9 3.00 × 10-2, 3.01 × 10-2
pip (MeOH)
morphc (MeOH) 138.8 9.04 × 10-3, 9.13 × 10-3
morph (MeOH)
morph (MeOH)
guad (EtOH)
gua (EtOH)
150.6 1.81 × 10-2, 1.82 × 10-2
156.8 2.48 × 10-2, 2.65 × 10-2
28.2 4.05 × 10-3, 4.06 × 10-3
38.4 6.10 × 10-3, 6.19 × 10-3
48.6 9.08 × 10-3
Resu lts a n d Discu ssion
gua (EtOH)
anile (MeOH)
thiouf (MeOH)
morph (MeOH)
morph (MeOH)
morph (MeOH)
gua (MeOH)
gua (MeOH)
gua (MeOH)
gua (MeOH)
100
130
ca. 20% reacn after 72 h
ca. 25% reacn after 72 h
It is generally accepted that activated SN2Ar reactions
of halogenoarenes, inter alia, proceed via a σ-complex
intermediate (Meisenheimer complex2,17 ) with flanking
transition states and that in the majority of such
reactions the formation of the first transition state is
rate-limiting.2,18,19 All available evidence suggests that
the SN2Ar reactions of coordinated halogenoarenes
follow the same reaction path.4a,c,d,5a,d,7c The nucleo-
philic reactivity order is determined by both Coulombic
and covalent contributions, whose relative contribution
depends on the initial charge distribution in the halo-
genoarene and the position of the transition state along
the reaction coordinate.
30.0 1.52 × 10-2
40.2 2.92 × 10-2
50.5 5.33 × 10-2
0.2 3.91 × 10-3, 3.61 × 10-3
10.0 7.01 × 10-3
20.1 1.52 × 10-2
32.9 4.30 × 10-2, 4.40 × 10-2
a
Standard deviation in calculating k2 was less than 3%, and
b
the correlation coefficient was not less than 0.998. Piperidine.
c Morpholine. Guanidine. e Aniline. Thiourea.
d
f
Ra te Da ta . The rate data for the reactions studied
in this work are presented in Table 1, while Figure 1
shows Arrhenius plots for all the reactions studied.
Table 2 presents the derived kinetic parameters and the
rate coefficients calculated at 0 °C for the reactions of
1. A comparison of the values of ∆Gq shows that in
reactions with 1 the anionic nucleophiles, apart from
N3-, are more reactive than pip and morph. The neutral
(8) See: Crampton, M. R. In Organic Reaction Mechanisms; Knipe,
A. C., Watts, W. E., Eds.; Wiley: New York, 1991, and previous
volumes.
(9) (a) Beckwith, A. L.; Miller, J .; Leahy, G. D. J . Chem. Soc. 1952,
3552. (b) Miller, J .; Wong, K. W. Aust. J . Chem. 1965, 18, 117. (c) Miler,
J .; Sakazaki, E. Unpublished work.
(10) Leahy, G. D.; Liveris, M.; Miller, J .; Parker, A. J . Aust. J . Chem.
1956, 9, 382.
(11) Lok, C. T.; Miller, J .; Stansfield, F. J . Chem. Soc. 1964, 1213.
(12) Bunnett, J . F.; Merritt, W. D. J . Am. Chem. Soc. 1957, 79, 5967.
(13) Coniglio, B. O.; Giles, D. E.; McDonald, W. R.; Parker, A. J . J .
Chem. Soc. B 1966, 152.
F igu r e 1. Arrhenius plots for reactions of the (η6-chlo-
robenzene)(η5-cyclopentadienyl)iron(III) cation with PhO-,
MeS-, PhS-, N3-, piperidine, morpholine, and guanidine.
(14) Bunnett, J . F.; Garbisch, E. W.; Pruitt, K. M. J . Am. Chem.
Soc. 1957, 79, 385.
(15) Chapman, N. B.; Parker, R. E. J . Chem. Soc. 1951, 3301.
(16) Miller, J .; Yeung, H. W. J . Chem. Soc., Perkin Trans. 2 1972,
1553.
(17) For reviews of Meisenheimer complexes see: (a) Terrier, F.
Chem. Rev. 1982, 82, 77. (b) Artamkina, G. A.; Egorov, M. P.;
Beletskaya, J . P. Chem. Rev. 1982, 82, 427. (c) Buncel, E.; Crampton,
M. R.; Strauss, M. J .; Terrier, F. Electron Deficient Aromatic- and
Heteroaromatic-Base Interactions; Elsevier: Amsterdam, Oxford, New
York, Tokyo, 1984.
nucleophile gua is also an exception, showing a reactiv-
ity similar to that of the anionic nucleophiles MeS- and
PhS-. The values of ∆Hq and ∆Sq show, as expected,
that the differences in reactivity between the anionic
and neutral nucleophiles are mainly entropic. Again,
the exceptions are N3-, which has abnormally high
values of entropy and enthalpy of activation, and gua,
which has an exceptionally low enthalpy of activation
while the entropy of activation is highly negative.
(18) Crampton, M. Adv. Phys. Org. Chem. 1969, 7, 211.
(19) Ritchie, C. D.; Sawada, M. J . Am. Chem. Soc. 1977, 99, 3754.
+
+

3200 Organometallics, Vol. 15, No. 14, 1996
Gomes et al.
Ta ble 2. Ra te Coefficien ts, Ca lcu la ted a t 0 °C, a n d Der ived Kin etic P a r a m eter s for Rea ction s of
(C6H5Cl)F e(C5H5)+ a
k2, 0 °C
∆Eq b
log Ab
∆Hq
∆Sq
∆Gq
nucleophile (solvent)
(dm3 mol-1 s-1
)
(kJ mol-1
)
(A, L mol-1 s-1
)
(kJ mol-1
)
(J mol-1 K-1
)
(kJ mol-1
)
MeO- (MeOH)
PhO- (MeOH)
MeS- (MeOH)
PhS- (MeOH)
N3- (MeOH)
pip (MeOH)
4.83 × 10-3
1.12 × 10-5
1.23 × 10-3
4.10 × 10-4
1.90 × 10-9
3.37 × 10-7
3.08 × 10-9
1.09 × 10-3
81.5
13.3
88.0 ( 0.9
11.9 ( 0.2
10.6 ( 0.2
11.8 ( 0.2
15.0 ( 0.7
8.1 ( 0.1
8.7 ( 0.2
3.13 ( 0.05
85.5
68
76.6
121
73.5
82
-26.2
-50.1
-28.5
30.8
-98.0
-105
-193
93.3
83
71 ( 1
79 ( 1
85.1
112
103
114
87.0
124 ( 5
76.0 ( 0.7
85 ( 2
morph (MeOH)
gua (EtOH)
31.9 ( 0.3
29.4
Values with MeO-/MeOH from ref 5b. The standard deviations were obtained from the experimental data of Table 1.
a
b
Ta ble 3. Ra te Coefficien ts, k2, a t 0 °C, a n d Ra tios k2(Nu )/k2(MeO-) a n d k2(1)/k2(2), a t 0 °C, for C6H3(NO2)2Cl
(2) a n d (C6H5Cl)F e(C5H5)+ (1)
C6H5(NO2)2Cl
(C6H5Cl)Fe(C5H5)+
k2 (dm3 mol-1 s-1 k2(Nu)/k2(MeOH-)
nucleophile (solvent)
k2 (dm3 mol-1 s-1)
k2(Nu)/k2(MeOH-)
)
k2(1)/k2(2)
MeO- (MeOH)
PhO- (MeOH)
MeS- (MeOH)
PhS- (MeOH)
N3- (MeOH)
pip (MeOH)
morph (MeOH)
gua (EtOH)
anil (EtOH)
2.00 × 10-3a
4.89 × 10-5b
3.66 × 10-3c
3.90d,e
1.0
4.83 × 10-3
1.12 × 10-5
1.2 × 10-3
4.1 × 10-4
1.7 × 10-9
3.8 × 10-7
3.1 × 10-8
1.09 × 10-3
1.0
2.42
2.4 × 10-2
2.3 × 10-3
2.55 × 10-1
8.5 × 10-2
3.5 × 10-7
7.8 × 10-5
6.4 × 10-6
2.3 × 10-1
2.3 × 10-1
3.3 × 10-1
1.1 × 10-4
5.1 × 10-5
1.9 × 10-4
1.8 × 10-5
3.1 × 10-1
1.8
2.0 × 103
1.7 × 10-2
9.8 × 10-1
8.5 × 10-1
1.8
3.31 × 10-5f
1.95 × 10-3g
1.70 × 10-3
3.50 × 10-3
1.08 × 10-5h
7.1 × 10-8e
5.4 × 10-3
20% reacn, 72 h, 100 °C
25% reacn, 72 h, 130 °C
thiou (MeOH)
3.5 × 10-5
a
b
d
g
h
Reference 9. Reference 10. c Reference 11. Reference 12. e Reference 16. f Reference 13. Reference 14. Reference 15.
Ta ble 4. Absolu te Va lu es ∆Eq (k J m ol-1) a n d log A
(A, d m 3 m ol-1 s-1) a n d Rela tive Va lu es Rela ted to
Rea ction w ith MeO-/MeOH ∆∆Eq (k J m ol-1) a n d ∆
log A of th e Ar r h en iu s P a r a m eter s for Rea ction s
of C6H3(NO2)2Cl a n d (C6H5Cl)F e(C5H5)+
Evidence exists that the cationic substrate 1 readily
forms an ion pair with MeO-, resulting in reduction of
its reactivity toward nucleophilic attack.7c The reduc-
tion in the second-order rate constant is ca. 10 times
when the concentration of MeO- increases by ca. 100
times. From the rate coefficients shown in Table 3, one
finds that MeO- is ca. 2 times more reactive toward the
cationic substrate 1 when compared with the neutral
substrate 2, in spite of ion pairing with 1. The same
comparison with PhS- shows that it is ca. 104 less
reactive toward 1. Also from Table 3 one finds that the
reactivities of PhO- and MeS- toward 1 are 0.2 and 0.3
times its reactivities toward 2, respectively. The reduc-
tion of the relative reactivity of PhS- compared to the
other anionic nucleophiles seems too pronounced to be
caused solely by a long-distance interaction such as ion
pairing, and some additional explanation is required.
C6H3(NO2)2Cl
∆Eq ∆∆Eq log A ∆ log A ∆Eq ∆∆Eq log A ∆ log A
(C6H5Cl)Fe(C5H5)+
nucleophile
(solvent
MeO- (MeOH) 73.0a
0
11.2
0
81.5
0
13.3
0
PhO- (MeOH) 78.5b +5.5 10.7 -0.5
MeS- (MeOH) 59.0c -14.0 8.9 -2.3
PhS_ (MeOH) 43.1d,e -29.9 8.9 -2.3
88.0 +6.5 11.9
-1.4
-2.7
-1.5
+1.7
71 -10.5 10.6
79
-2.5 11.8
N3- (MeOH)
pip (MeOH)
morph (MeOH) 49.4 -23.6 6.7 -4.5
gua (EtOH)
anil (EtOH)
74.9f
+1.9 9.9 -1.3 124 +42.5 15.0
48.6g -24.4 6.7 -4.5
76.0 -5.5 8.10 -5.2
85 -4.6
31.9 -49.6 3.13 -10.2
+3.5 8.7
52.2 -20.8 7.5 -3.7
46.5h -26.5 4.0 -7.2
thiou (MeOH) 76.6e +3.6 7.5 -3.7
a
b
d
Reference 9. Reference 10. c Reference 11. Reference 12.
e Reference 16. f Reference 13. Reference 14. Reference 15.
g
h
An explanation based on the position of the transition
state along the reaction coordinate could be advanced
when comparing the reactivities of MeO- and MeS-
toward 1 and 2. In reactions with 1 the hard nucleo-
phile MeO- is 4 times more reactive than the soft
nucleophile MeS-. This order is reversed for 2, which
reacts with MeS- almost twice as fast as with MeO-.
This comparison suggests that in reactions of 1 with
these anionic nucleophiles there might be a tendency
toward greater Coulombic contributions and less cova-
lent contribution to the transition state20 in comparison
to the corresponding reactions of 2. In other words,
there might be a tendency toward earlier transition
states with 1. When the reactivities of PhS- and MeO-
toward 1 and 2 are compared, PhS- is ca. 2 × 103 times
more reactive than MeO- 12,21 with 2, whereas, in
complete contrast, it is 8.5 × 10-2 as reactive with 1.
This is a change of well over 104 in relative reactivities.
Table 3 shows the rate coefficients, at 0 °C, for the
reactions of 1 and 2 compared with MeO- as the
standard nucleophile, and Table 4 shows the absolute
and relative kinetic parameters for these reactions.
From the values of k2 at 0 °C, the following order of
reactivities toward 1 is apparent for the anionic nucleo-
philes: N3- << PhO- < PhS- < MeS- < MeO-. These
nucleophiles show a completely different order of reac-
tivity toward 2, viz., N3- < PhO- < MeO- < MeS- <<
PhS-. The order of reactivity of the neutral nucleophiles
is the same for both substrates, that is, thiou < anil
<< morph < pip < gua.
We now examine these differences in reactivity in
more detail.
Rea ction s of MeO-, MeS-, P h O-, a n d P h S-. The
relative reactivities of these anionic nucleophiles with
the neutral substrate 2 and cationic substrate 1 might
be affected by at least three factors: (i) ion pairing of
the nucleophile with the iron center in 1, (ii) the position
of the transition state along the reaction coordinate, and
(iii) stereoelectronic factors in the transition state for
nucleophilic attack.
(20) See: Hudson, R. F. In Chemical Reactivity and Reaction Paths;
Klopman, G., Ed.; Wiley: New York, USA, 1974; p. 167.
(21) Bartoli, G.; Todesco, P. E.; Fiorentino, M. J . Am. Chem. Soc.
1977, 99, 6874.
+
+

SN2Ar Reactions of Free and Coordination Arenes
Organometallics, Vol. 15, No. 14, 1996 3201
factors of 7 and 2 × 104, respectively, on changing from
2 to 1. This suggests that it is repulsion between the
arenide electrons and the phenyl ring of PhS- that is
important.
On the other hand, the corresponding effects for the
ratio k2(PhO-)/k2(MeO-), compared with k2(PhS-)/k2-
(MeO-), is much smaller, a decrease of 101 instead of 2
× 104, and it can be concluded that the shorter bond
formed between PhO- and the reaction center precludes
the conformation in which the phenyl ring attached to
the oxygen lies above and close to the arenide electrons
in the ring undergoing substitution. The only repul-
sions are due to the lone-pair electrons on the oxygen
atoms, and these are evidently less important.
F igu r e 2. PhS-/MeO- rate coefficient ratios, in methanol,
at 0 °C: (a) ref 25; (b) this work.
This can hardly be ascribed exclusively to a greater
covalent contribution to the transition state in the
reactions with 2, though polarizability-polarizability
interactions22 seem to be important in determining the
thiolate reactivities toward aryl halides.11,23
Rea ction s of N3-. The azide anion shows an abnor-
mally low reactivity toward the cationic substrate 1 (k2-
(0 °C) ) 1.7 × 10-9 dm3 mol-1 s-1) when compared to
the other nucleophiles studied (ca. 10-3-10-5). It is also
much less reactive with 1 than with neutral 2 (k2(0 °C)
) 3.31 × 10-5 dm3 mol-1 s-1). These results are rather
A third explanation should also be considered, and it
turns out to be preferable. In general, the approach of
a nucleophile to a reaction center results in displace-
ment of substrate electrons from the reaction center to
other locations in the substrate. The site of this
relocation and its effects are likely to be different with
different substrates. Thus, it has been suggested24 that,
in the reactive intermediates formed by 2, electrons
have been displaced onto the â-exocyclic oxygen atoms
of the two nitro groups. (These displaced electrons were
conveniently called arenide electrons.24 ) This displace-
ment is relatively facile. On the other hand, the marked
reduction in the relative nucleophilicities of PhS-
toward a series of heterocyclic arenes was attributed24
to the fact that the arenide electrons are now located
mainly in the heteroarene ring itself. As a consequence,
an entering group such as PhS-, with its substantial
steric and electronic requirements, will suffer consider-
able repulsion from the arenide electrons on the nearby
ring when forming the reaction intermediates. On the
basis of the values of the ratios k2(PhS-)/k2(MeO-) for
reactions with the substrates shown in Figure 2, we can
conclude that this stereoelectronic repulsion effect is
operating in the SN2Ar reactions of 1, to an even greater
extent than in reactions of the heteroarenes. This
repulsion occurs in spite of the transfer of some π-elec-
tron density from the C6H5Cl ligand to the Fe2+ center.
It could involve repulsion between the arenide electrons
and either the π-electrons on the phenyl ring of PhS-
(3a ) or the lone pairs on the S atom in PhS- (3b), the
-
surprising, since N3 reacts with carbocations25 much
faster than with 2.26 The enthalpy of activation (121
kJ mol-1) is much higher than those of the other anionic
nucleophiles (see Table 2).
A possible explanation for this surprisingly low
-
reactivity of N3 toward 1 involves incipient coordina-
tion of the azide ion to the Fe2+ center, stabilizing the
ground state. The highly positive entropy of activation
(33 J K-1 mol-1) favors this hypothesis.
Rea ction s of p ip , m or p h , a n d a n il. The question
of ion pairing in the reactivities of these neutral nu-
cleophiles with 1 does not arise. In the reactions of 1
with these amines, greater covalent contribution to the
rate-limiting transition state is expected compared with
reactions of the anionic nucleophiles. The formation of
the new bonds is accompanied by the development of
formal positive charge in the nucleophile. This is
consistent with the entropies of activation that are much
more negative for the neutral nucleophiles compared
with those of anionic nucleophiles (see Table 2).
The larger contribution of the new covalent bonds
leads to more stringent geometrical requirements in the
reaction intermediates. With pip, morph, and anil the
short C-N bonds and the multivalent character of
nitrogen are likely to lead to adverse interactions, which
become still more effective when the arenide electrons
remain in the electrophilic ring. This could be respon-
sible for the marked changes in relative reactivities of
pip, morph, and anil toward 1 and 2.
In the reaction of pip and morph with 2 the amines
show roughly the same reactivity as MeO-, viz. k2-
(amine)/k2(MeO-) ≈ 1 and 0.9, respectively. However,
the changes in the Arrhenius parameters are large. The
values of ∆Eq are much smaller for the amines with
∆∆Eq ) ∆Eq(amine) - ∆Eq(MeO-) of ca. -24 kJ mol-1
.
This is slightly overcompensated by the low values of
log A (∆ log A ) -4.5 for both amines). In the reactions
with cation 1 the rate coefficient ratios k2(amine)/k2-
(MeO-) are much smaller, viz. 7.8 × 10-5 and 6.4 × 10-6
for pip and morph, respectively. Associated with this,
the larger values of ∆Eq, with ∆∆Eq values of -5.5 kJ
mol-1 for pip and +3.5 kJ mol-1 for morph, are note-
worthy. The difference among the amines (k2(pip)/k2-
σ-complex being used to represent the first, and rate-
determining, transition state of the reactions. If the
ratios k2(MeS-)/(MeO-) and k2(PhS-)/k2(MeO-) in Table
3 are considered, it can be seen that they decrease by
(22) Bunnett, J . F.; Nudelman, N. S. J . Org. Chem. 1969, 34, 2038.
(23) Bunnett, J . F.; Kato, T.; Nudelman, N. S. J . Org. Chem. 1969,
34, 785.
(24) Cito, A. M. G. C.; Lopes, J . A. D.; Miller, J .; Moran, P. J . S. J .
Chem. Res., Synop. 1983, 184; J . Chem. Res., Miniprint 1983, 1586.
(25) Bunton, C. A.; Moffat, J . R.; Rodenas, E. J . Am. Chem. Soc.
1982, 104, 2653.
(26) Ritchie, C. D. J . Am. Chem. Soc. 1975, 97, 1170.
+
+

3202 Organometallics, Vol. 15, No. 14, 1996
Gomes et al.
(morph) ≈ 10) possibly involves greater repulsion to the
arenide electrons when an oxygen in morph is substi-
tuted for the 4-CH2 group in pip. The former, with its
unshared electrons and fractional negative charge, can
reasonably be assumed to repel the arenide electrons
more.
nitrogen atoms to counteract unfavorable interactions
of the lone pairs of electrons on the sulfur atom with
the arenide electrons in the transition state. With 2,
k2(thiou)/k2(MeO-) is 3.5 × 10-5 with ∆∆Eq ) 3.6 kJ
mol-1 and ∆ log A ) -3.7. We have a crude quantita-
tive result for reaction with 1, viz. about 25% reaction
in 72 h at 130 °C. From this we estimate a drop in
reactivity relative to MeO- of ca. 100 times in contrast
with the drop of only ca. 4 times observed for gua.
Aniline is significantly less reactive toward 2 than
MeO-, pip, and morph, and the rate-coefficient k2(anil)/
k2(MeO-) is 5.4 × 10-3. As with the other amines, the
value of ∆Eq is low and ∆∆Eq is -26.5 kJ mol-1. This
effect is overcompensated by a very low value of log A,
corresponding to ∆ log A ) -7.2. Only about 20% of
the reaction with 1 had occurred after 72 h at 100 °C.
From this we estimated, very crudely, a drop in the
reactivity compared with 2 of the order of 104, similar
to the more precise values obtained for pip and morph.
Rea ction s of gu a a n d th iou . The behavior of
guanidine is completely different from that of the other
amines, differing also in belonging to the class of
unsaturated nucleophiles of the type ZsCdNu. The
internal conjugation in 4 increases the nucleophilic
Exp er im en ta l Section
Ma ter ia ls. Chlorobenzene, (η5-cyclopentadienyl)iron(II),
sodium tetrafluoroborate, sodium azide, piperidine, morpho-
line, guanidine, aniline, thiourea, phenol, and hydrogen
phthalate were commercial reagents and were recrystallized
under argon, or sublimed under argon, as appropriate. Melt-
ing points and boiling points of all organic reagents agreed
with those reported in the literature.
(η6-Chlorobenzene)(η5-cyclopentadienyl)iron tetrafluorobo-
rate was prepared as described in the literature,5b,27 without
important modifications. Methanol was dried over Mg and I2
and distilled, then refluxed over AgNO3, and distilled again.
The fraction collected distilled between 63.5 and 64.8 °C.
Ethanol was dried over CaO, refluxed over AgNO3, then over
KOH, and finally over diethyl phthalate, and distilled. The
fraction used was collected at 78 °C. Sodium methoxide in
methanol was prepared from purified methanol and previously
cleaned sodium and stored under dinitrogen, with protection
against entrance of CO2. Its concentration was determined
periodically by standard titrimetric procedures, against hy-
drogen phthalate. Sodium phenoxide in methanol was ob-
tained from phenol and methanolic sodium methoxide.
Ap p a r a tu s. A Metrohm-Herisau automatic titrator was
used in all the chloride titrations and pH measurements. In
the chloride titrations, silver chloride/saturated calomel con-
jugated electrodes were used. In the pH measurements glass
and saturated calomel electrodes were used. The 1H NMR
spectra were obtained on a Varian T-60 spectrometer, using
TMS as internal standard. The IR spectra were recorded on
a Perkin-Elmer 283 spectrometer, using KBr pellets. Elemen-
tal analyses were carried out using a Perkin-Elmer 2400 CHN
elemental analyzer.
Gen er a l P r oced u r e for Isola tion of Rea ction P r od u cts.
Substitution products were isolated by recrystallization of the
picrate salts. (η6-Chlorobenzene)(η5-cyclopentadienyl)iron(II)
tetrafluoroborate was dissolved in methanol (or ethanol) and
treated with the nucleophile. After the reaction was completed,
the reaction mixture was filtered and evaporated to dryness
under vacuum below 40 °C. The residue was dissolved in
water, excess base was neutralized with 10% aqueous HCl,
and the product was precipitated by addition of a saturated
aqueous solution of picric acid. It was then filtered by suction,
dried under vacuum, and recrystallized from ethanol. The
products were characterized by 1H NMR and infrared spectra.
Elemental analyses of the stable products were in good
agreement with calculated values. 1H NMR spectra of previ-
ously known compounds were identical with those reported.
The data are as follows.
strength of the NH group. In addition to this, gua has
much smaller steric demand than pip and morph.
Guanidine thus shows a reactivity toward 2 ca. 2 times
greater than MeO- compared with the rate-coefficient
ratios (1 and 0.9) observed for pip and morph. This
value is related to a drop in the activation energy (∆∆Eq
) ∆Eq(gua) - ∆Eq(MeO-)) of ca. 21 kJ mol-1 which is
compensated for by a drop of 3.7 in log A.
In the reactions with 1 the results show a complete
contrast with the other amines, its reactivity being
comparable to that of MeO-. Thus, k2(gua)/k2(MeO-)
) 0.23, which is about 3 and 4 orders of magnitude
greater than the values for pip and morph, respectively.
Arrhenius parameters show reductions in ∆Eq of 50 kJ
mol-1 compared with MeO-, instead of -5.5 kJ mol-1
for pip and +3.5 kJ mol-1 for morph, and a drop in log
A of 10 instead of ca. 5 for pip and morph. This can be
interpreted in terms of the development of positive
charge delocalized over the terminal nitrogen atoms
when the saturated nitrogen atom reacts with the
substrate leading to the intermediate 5. It is reasonable
to suppose that the favorable electrostatic interaction
is responsible for the stabilization of the transition state.
(η6-Meth oxyben zen e)(η5-cyclop en ta d ien yl)ir on (II) P i-
cr a te. Following the general procedure, 0.320 g (1 mmol) of
1 reacted with 0.054 g (1 mmol) of NaOCH3 in 10 cm3 of
methanol, for 3 h, at room temperature, yielding 0.343 g (75%)
of pale yellow needles; mp 273 °C. 1H NMR (in DMSO-d6): δ
8.60 (s, 2H, C6H2N3O7), 6.00 (m, 5H, C6H5), 4.80 (s, 5H, C5H5),
2.80 (s, 3H, CH3O). IR (KBr): 470 (iron-ring stretching), 490
cm-1 (ring tilt). Anal. Calcd for C18H15FeN3O8: C, 47.29; H,
3.31; N, 9.19. Found: C, 47.33; H, 3.30, N, 9.40.
Finally, we consider the reaction of thiou, another
ZsCdNu type nucleophile. Results for thiou and a
series of methylated thioureas9c suggest that the inter-
nal conjugation has relatively little importance, so that
their nucleophilicity is mainly due to the intrinsic
behavior of the CdS group. As a consequence, there is
relatively little development of positive charge on the
(η6-(Met h ylt h io)b en zen e)(η5-cyclop en t a d ien yl)ir on -
(II) P icr a te. Following the general procedure, 0.320 g (1
mmol) of 1 reacted with 0.070 g (1 mmol) of NaSCH3 in 10
(27) Federman Neto, A.; Miller, J . An. Acad. Bras. Cienc. 1982, 54,
331.
+
+

SN2Ar Reactions of Free and Coordination Arenes
Organometallics, Vol. 15, No. 14, 1996 3203
cm3 of methanol, for 3 h, at room temperature, yielding 0.317
g (67%) of deep yellow needles sensitive to air and light; mp
170 °C dec. 1H NMR (in DMSO-d6): δ 8.55 (s, 2H, C6H2N3O7),
6.20 (m, 5H, C6H5), 4.95 (s, 5H, C5H5), 3.15 (s, 3H, CH3S; IR
(KBr): 450 (iron-ring stretching), 490 cm-1 (ring tilt). Anal.
Calcd for C18H15FeN3O7S: C, 45.67; H, 3.20; N, 8.88. Found:
C, 45.65; H, 2.84; N, 9.06.
(η6-P h en oxyben zen e)(η5-cyclop en ta d ien yl)ir on (II) P i-
cr a te. Following the general procedure, 0.320 g (1 mmol) of
1 reacted with a mixture of 0.054 g (1 mmol) of NaOCH3 and
0.24 g (2.5 mmol) of C5H5OH28 in 10 cm3 of methanol for 3 h
under reflux,29 yielding 0.321 g (62%) of pale yellow needles;
mp 210-214 dec. 1H NMR (DMSO-d6): δ 8.40 (s, broad, 2H,
C6H2N3O7), 6.00-6.70 (m, 10H, C6H5OC6H5), 5.10 (s, broad,
5H, C5H5). IR (KBr): 660 (bending, C-C aromatic), 470 (iron-
ring stretching), 485 cm-1 (ring tilt). Anal. Calcd for
cm3 of methanol, for 2 h under reflux, yielding 0.323 g (63%)
of orange needles; mp 157-159 °C dec. 1H NMR (DMSO-d6):
δ 8.40 (s, 2H, C6H2N3O7), 6.90 (m, 5H, C6H5), 5.00 (s, 5H, C5H5),
3.00-3.40 (m, 8H, C4H8NO). IR (KBr): 2980-2800 (C-H
stretching), 470-460 (skeletal C-C deformation on N-alkyl
groups), 440 (ring tilt), 420 cm-1 (iron-ring stretching). Anal.
Calcd for C21H20FeN4O8: C, 49.21; H, 3.94; N, 10.94. Found:
C, 49.42; H, 3.84; N, 10.47.
Some of the organometallic cations reported in this paper
have previously been reported in the literature with other
counterions, viz. (η6-methoxybenzene)(η5-cyclopentadienyl)-
iron(II) as the tetrafluoroborate and hexafluorophosphate salts,
(η6-(phenylthio)benzene)(η5-cyclopentadienyl)iron(II) as the
tetraphenylborate salt,5b and (η6-phenoxybenzene)- and (η6-
azidobenzene)(η5-cyclopentadienyl)iron(II) as the hexafluoro-
phosphate salts.5e
C
23H17FeN3O8: C, 53.18; H, 3.30; N, 8.09. Found: C, 53.45;
Kin etics. The reactions at temperatures below 45 °C were
carried out in thermostated standard volumetric flasks, under
a purified dinitrogen atmosphere. Aliquots measured with
calibrated pipettes, withdrawn from reaction solutions, were
added to an excess of diluted nitric acid, and the concentrations
of the displaced chloride ions were measured. The reactions
at temperatures above 45 °C were carried out with samples,
previously measured with calibrated pipets, sealed in glass
ampules. In both procedures the chloride ions displaced by
the reacting nucleophile were titrated potentiometrically
against standard 0.05 M AgNO3 solution. The reactions of (η6-
chlorobenzene)(η5-cyclopentadienyl)iron(II) tetrafluoroborate,
as substrate, with sodium methoxide and sodium azide in
methanol were carried out with equimolar concentrations of
reagent and substrate. To prevent the interference of meth-
oxide ions from the solvent, the reactions of the substrate with
benzenethiolate, methanethiolate, and phenoxide ions in
methanol were carried out with equimolar concentrations of
reagents and their conjugated acids, as well as the substrate.
For the same reason the reactions of the substrate with aniline,
morpholine, and thiourea in methanol and with guanidine in
ethanol were carried out with 2 equiv of the conjugated acid,
2 equiv of the free base, and 1 equiv of the substrate. The
initial concentration of the substrate in all the kinetic runs
was 0.025 M.
H, 2.97; N, 7.87.
(η6-(P h en ylt h io)b en zen e)(η5-cyclop en t a d ien yl)ir on -
(II) P icr a te. Following the general procedure, 0.320 g (1
mmol) of 1 reacted with a mixture of 0.054 g (1 mmol) of
NaOCH3 and 0.13 g (1.2 mmol) of C6H5SH30 in 10 cm3 of
methanol, for 3 h, at room temperature, yielding 0.342 g (64%)
of deep yellow crystals sensitive to air and light); mp 120 °C
dec. 1H NMR (DMSO-d6): δ 8.45 (s, 2H, C6H2N3O7), 5.90-
6.50 (m, broad, 10H, C6H5SC6H5), 5.00 (s, 5H, C5H5). IR
(KBr): 450 (iron-ring stretching), 480 cm-1 (ring tilt). Anal.
Calcd for C23H17FeN3O7S: C, 51.59; H, 3.20; N, 7.85. Found:
C, 51.23; H, 2.84; N, 7.83.
(η6-Azidoben zen e)(η5-cyclopen tadien yl)ir on (II) P icr ate.
Following the general procedure, 0.320 g (1 mmol) of 1 reacted
with 0.065 g (1 mmol) of NaN3 in 20 cm3 of methanol31 for 2 h,
under reflux, yielding 0.257 g (55%) of yellow needles very
sensitive to light; mp 156-158 °C. 1H NMR (CH3CN-d6): δ
8.60 (s, 2H, C6H2N3O7), 5.80-6.50 (m, 5H, C6H5N), 5.00 (s, 5H,
C5H5). IR (KBr): 2030-2150 (N-N-N stretching), 640 (N-
N-N bending), 470-480 cm-1 (ring tilt and iron-ring stretch-
ing). Anal. Calcd for C17H12FeN6O7: C, 43.59; H, 2.58; N,
17.95. Found: C, 43.08; H, 2.46; N, 17.27.
(η6-P iper idin oben zen e)(η5-cyclopen tadien yl)ir on (II) P i-
cr a te. Following the general procedure, 0.320 g (1 mmol) of
1 reacted with 2.3 g (27 mmol) of piperidine32 in 10 cm3 of
methanol for 4 h at room temperature, yielding 0.301 g (59%)
of orange needles, sensitive to light; mp 110-113 °C dec. 1H
NMR (DMSO-d6): δ 8.50 (s, 2H, C6H5N3O7), 6.00 (m, 5H, C6H5),
4.90 (s, 5H, C5H5), 3.00-3.60 (m, 10H, C5H10N). IR (KBr):
2850, 2915 (C-H stretching), 420-520 cm-1 (skeletal C-C
deformation on N-alkyl groups, iron-ring stretching, and ring
tilt). Anal. Calcd for C22H22FeN4O7): C, 51.76; H, 4.35; N,
10.98. Found: C, 51.23; H, 4.28; N, 10.86.
The values of the rate coefficients were obtained by the
method of least squares from a linear correlation of 1/(a - x)
versus time in seconds. In the case of reactions with aniline,
morpholine, piperidine, thiourea, and guanidine the values of
k2 were obtained from a linear correlation of 1/{2(a - x)} versus
time in seconds. The Arrhenius and derived kinetic param-
eters were obtained by the same procedure, from a linear
correlation of log k2 (rate coefficients in dm3 mol-1 s-1) versus
reciprocal temperature, 1/T.
(η6-Mor p h olin ob e n ze n e )(η5-cyclop e n t a d ie n yl)ir on -
(II) P icr a te. Following the general procedure, 0.320 g (1
mmol) of 1 reacted with 0.174 g (2 mmol) of morpholine in 10
We did not carry out a complete study of the reaction with
methoxide ion, since this had been studied by Nesmeyanov et
al.5b However, as a precaution, we have determined the k2
value at 13.6 °C as 2.41 × 10-2 dm3 mol-1 s-1 with a correlation
coefficient of 0.9996. Nesmeyanov’s data5a lead to the value
2.485 × 10-2 dm3 mol-1 s-1, in satisfactory agreement.
(28) Excess phenol is used to avoid possible competition from
methoxide.
(29) No reaction is observed at room temperature.
(30) Excess thiophenol is used to avoid possible competition from
methoxide.
(31) A larger volume of methanol is used compared to the other
reactions because NaN3 is less soluble in methanol than the other
reagents.
(32) If a smaller excess of piperidine is used, the methoxide product
is obtained.
Ack n ow led gm en t. E.J .S.V. thanks Prof. Anthony
J . Poe and Mr. J ianbin Hao for their help in the final
edition of the manuscript.
OM9601817
+
+