Kinetics and mechanism of the aminolysis of O-ethyl S-aryl thiocarbonates in acetonitrile

Article abstract of DOI:10.1002/(SICI)1097-4601(2000)32:3<131::AID-KIN2>3.0.CO;2-C

The Kinetics and mechanism of the reactions of O-ethyl S-(Z)aryl thiocarbonates with (X)benzylamines in acetonitrile at 45.0 °C are studied. Relatively small values of β_x (β_nuc) = 0.6 to approximately 0.8 and β_z (β_lg) = -0.5 to approximately -0.7 together with a negative cross-interaction constant ρ_xz (= -0.47) and failure of the reactivity-selectivity principle (RSP) are interpreted to indicate a concerted mechanism. The normal kinetic isotope effects (k_H/k_D = 1.3 to approximately 1.8) involving deuterated benzylamine nucleophiles suggest a hydrogen-bonded, four-center-type transition state.

Full text of DOI:10.1002/(SICI)1097-4601(2000)32:3<131::AID-KIN2>3.0.CO;2-C

Kinetics and Mechanism of
the Aminolysis of O-Ethyl
S-Aryl Thiocarbonates
in Acetonitrile
HYUK KEUN OH,1 YUN HO LEE,1 IKCHOON LEE2
1Department of Chemistry, Chonbuk National University, Chonju, 560-756, Korea
2Department of Chemistry, Inha University, Inchon 402-751, Korea
Received 22 April 1999; accepted 21 October 1999
ABSTRACT: The kinetics and mechanism of the reactions of O-ethyl S-(Z)aryl thiocarbonates
with (X)benzylamines in acetonitrile at 45.0ЊC are studied. Relatively small values of ␤X
(␤nuc) ϭ 0.6 ϳ 0.8 and ␤Z (␤lg) ϭ Ϫ0.5 ϳ Ϫ0.7 together with a negative cross-interaction
constant ␳XZ (ϭ Ϫ0.47) and failure of the reactivity–selectivity principle (RSP) are interpreted
to indicate a concerted mechanism. The normal kinetic isotope effects (kH/kD ϭ 1.3 ϳ 1.8)
involving deuterated benzylamine nucleophiles suggest a hydrogen-bonded, four-center-type
transition state. ᭧ 2000 John Wiley & Sons, Inc. Int J Chem Kinet 32: 131–135, 2000
INTRODUCTION
O
S
R9C9S9Ar
R9C9O9Ar
The aminolyses of acyl compounds have been studied
extensively. Bro¨nsted plots were used in these reac-
tions as mechanistic criteria [1,2]. In many of these
nucleophilic reactions curved Bro¨nsted-type plots
have been found, which have been attributed to a
change in the rate-determining stepfrom breakdown
(␤nuc Х 0.8 ϳ 1.0) to formation (␤nuc Х 0.1 ϳ 0.3) of
a tetrahedral intermediate, TϮ, in the reaction path as
the basicity of the amine nucleophile increases [1–6].
The change in slope occurs at pKa, where the leaving
groupand the nucleophile in the intermediate have the
same leaving ability. Quite interestingly, however,
concerted processes are found only in the reactions
between O-ethyl S-aryl-thiocarbonates (structure I
with R ϭ EtO) with good leaving groups (Ar ϭ 2,4-
(NO2)2- and 2,4,6-(NO2)3-C6H2) and alicyclic second-
ary amines (␤nuc ϭ 0.4 ϳ 0.6) [7,8]. The concerted
mechanism has been shown to be enforced by (i) in-
stability incurred by the ArS groupbecause ArS Ϫ is
Structure I
Structure II
S
R9C9S9Ar
Structure III
much less basic than ArOϪ and a better nucleofuge
from TϮ than ArOϪ group [4]; (ii) stronger push pro-
vided by EtO, which enhances the nucleofugality of
both the amine and arylthiolate ion from the interme-
diate relative to other R groups [9–11] (e.g., phenyl
and alkyl); (iii) much faster expulsions of a given
amine and ArSϪ from TϮ formed with structure I than
those from TϮ with structures II and/or III due to a
stronger ␲-bonding energy of the carbonyl groupcom-
pared with thiocarbonyl [9]; and (iv) greater “push” to
expel ArSϪ from TϮ provided by the secondary (ali-
cyclic) amines than the tertiary (pyridines) amines
[11].
Correspondence to: I. Lee
᭧ 2000 John Wiley & Sons, Inc.
To examine further the driving force for the con-

132
OH ET AL.
1
1
certed aminolysis of the thiol derivatives, structure I,
we carried out kinetic studies of the aminolysis of
structure I with R ϭ EtO and Ar ϭ C6H4Z (Z ϭ
p-Me, H, p-Cl, and p-Br) using benzylamines
(XC6H4CH2NH2; X ϭ p-OMe, p-Me, H, p-Cl, and M-
Cl) in acetonitrile at 45.0ЊC. Another important objec-
tive of this work is to determine the cross-interaction
constant, ␳XZ in Eq. (1) [12–14] where X and Z denote
substituents in the nucleophile and leaving group, re-
spectively. It has been postulated and experimentally
found that in a stepwise acyl transfer
tonitrile. The second-order rate constants, k2 (MϪ sϪ ),
were obtained as
kobs ϭ k0 ϩ k2 [N]
(2)
the slopes of the plots of kobs vs. benzylamine concen-
tration [N] and are summarized in Table I. No third-
order or higher order terms in amine were detected and
no complications were found neither in the determi-
nation of kobs nor in the linear plots of Eq. (2). This
suggests that there is no base catalysis or noticeable
side reactions. The rate is faster with a stronger nucle-
ophile and a better nucleofuge as normally expected
log (kXZ/kHH) ϭ ␳ ␴ ϩ ␳ ␴ ϩ ␳ ␴ ␴
XZ X Z
(1)
X
X
Z
Z
from
a nucleophilic substitution reaction. The
Bro¨nsted ␤X (␤nuc) and ␤Z (␤lg) and Hammett ␳X (␳
)
nuc
through a tetrahedral intermediate the sign of ␳ is
XZ
and ␳Z (␳lg) values are also shown in Table I. We note
that the magnitudes of these selectivity parameters are
in general considerably smaller than those for the ami-
nolysis with benzylamines involving rate-limiting ex-
pulsion of the leaving group, ArSϪ, from a tetrahedral
intermediate, TϮ. For example, for the aminolysis of
thiolphenyl benzoates (structure I with R ϭ Ph) with
benzylamines in acetonitrile [11], which is believed to
proceed by a stepwise mechanism with rate-limiting
breakdown TϮ, the magnitude of ␳ (␤X) and ␳ (␤Z)
invariably positive and the reactivity–selectivity prin-
ciple (RSP) [15,16] holds [17,18]. In contrast, the sign
of ␳XZ is normally negative and the RSP does not hold
in the concerted acyl transfer reactions [17,18].
RESULTS AND DISCUSSION
The pseudo-first-order rate constants (kobs) for all re-
actions obeyed Eq. (2) with negligible k0 (Х0) in ace-
X
Z
values were much larger with Ϫ1.88 (1.86) and 3.84
Ϫ
Ϫ
Table I The Second-Order Rate Constants, k2 ϫ 103 dm3 mol 1 s 1 for the Reactions of O-Ethyl S-Aryl
Thiocarbonates with X-Benzylamines in Acetonitrile at 45.0ЊC
Z
a
b
X
p-Me
H
p-Cl
p-Br
␳
␤
Z
Z
p-OMe
18.4
7.51d
3.06e
31.7
75.4
85.6
38.7d
17.4e
1.63 Ϯ 0.09e
Ϫ0.66 Ϯ 0.03c
p-Me
H
16.8
26.9
21.8
65.1
50.0
71.8
1.56 Ϯ 0.11
1.51 Ϯ 0.07
Ϫ0.64 Ϯ 0.01
Ϫ0.63 Ϯ 0.02
13.2
53.1
9.70d
36.1d
4.81e
18.8e
p-Cl
2.35
7.80
Ϫ0.60 Ϯ 0.01c
0.62 Ϯ .02h
15.5
12.4
Ϫ0.63 Ϯ 0.01
0.63 Ϯ 0.02
31.6
25.2
Ϫ0.76 Ϯ 0.02
0.78 Ϯ 0.01
9.02
26.7
Ϫ0.78 Ϯ 0.01
0.79 Ϯ 0.03
1.37 Ϯ 0.09
1.32 Ϯ 0.05
Ϫ0.55 Ϯ 0.03
Ϫ0.54 Ϯ 0.03
m-Cl
f
␳
X
␳
g ϭ
Ϫ0.47 Ϯ 0.14c
XZ
h
␤
X
a The ␴ values were taken from Dean, J. A. Handbook of Organic Chemistry; McGraw-Hill: New York, 1987; Table 7-1. Correlation
coefficients were better than 0.995 in all cases.
b
The pKa values were taken from Buckingham, J., Ed.; Dictionary of Organic Chemistry, 5th ed. Chapman and Hall: New York, 1982;
Z ϭ p-Br was excluded from the Bro¨nsted plots for ␤Z due to an unreliable pKa values. Correlation coefficients were better than 0.998 in all
cases.
c Errors shown are standard deviations.
d
At 35ЊC.
e At 25ЊC.
f The ␴ values were taken from McDaniel, D. H.; Brown, H. C. J Org Chem 1958, 23, 420. Correlation coefficients were better than 0.998
in all cases.
g
Correlation coefficient was 0.997.
h
The pKa values were taken from Fischer, A.; Galloway, W. J.; Vaughan, J. J Chem Soc 1964, 3588. Correlation coefficients were better
than 0.997 in all cases. X ϭ p-CH3O was excluded from the Bro¨nsted plots for ␤X due to an unreliable pKa value listed.

KINETICS AND MECHANISM OF THE AMINOLYSIS OF O-ETHYL S-ARYL
133
(Ϫ1.63) for Z ϭ H and X ϭ H, respectively. These
are larger by ca. three times than those corresponding
values, Ϫ0.63 (0.63) and 1.51 (Ϫ0.63), in Table I. The
␤X value of 0.63 obtained in the present work is similar
to those for the concerted reaction of structure (R ϭ
EtO) with alicyclic (secondary) amines [7,8] (␤X ϭ
0.56 for Ar ϭ 2,4-(NO2)2C6H3 and ␤X ϭ 0.48 for
Ar ϭ 2,4,6-(NO2)2C6H2 in structure I) in aqueous so-
lution. These two derivatives of structure I (R ϭ EtO)
are, however, known to react with pyridines (tertiary
amines) by a stepwise mechanism with rate-limiting
breakdown of the intermediate, TϮ, with ␤X ϭ 0.9
(pKao ϭ 8.6) and ␤X ϭ 0.8 (pKao ϭ 7.3) for Ar ϭ 2,4-
(NO2)2- and 2,4,6-(NO2)3-C6H2S in structure I (R ϭ
EtO), respectively [19]. This means that the change of
amine from secondary to tertiary amines leads to an
increase in the magnitude of ␤X. On the other hand,
the aminolysis of O-ethlyl S-(Z-phenyl) dithiocarbon-
ate (structure III with R ϭ EtO and Ar ϭ C6H4Z) with
anilines in acetonitrile at 30.0ЊC was found to proceed
by a concerted mechanism (␤X ϭ 0.5 ϳ 0.7 and
TϮ, (i) by a strong electron releasing power of the R
group(R ϭ EtO has a stronger electron-releasing ef-
fect (␴R ϭ Ϫ0.50) than R ϭ Me (␴R ϭ Ϫ0.18)) [21];
(ii) by a strong “push” provided by a primary amine
in TϮ (the push provided by amines in the putative
intermediate, TϮ, decreases in the order, primary Ͼ
secondary Ͼ tertiary due to stabilization provided by
the cationic charge dispersion by the amines within
TϮ) [11]; (iii) by a relatively strong leaving ability of
the ArS group(lower p Ka than the corresponding ArO
group), and (iv) by a destabilizing effect of TϮ by the
solvent, acetonitrile [3].
The kinetic isotope effects (kH/kD) determined with
deuterated benzylamine nucleophiles are collected in
Table II. The kH/kD values are all substantially greater
than unity, suggesting that a four-center-type TS
(structure IV) is involved [14]. In agreement with the
negative ␳ [13], which can be alternatively defined
XZ
as Eq. (4), the magnitude of kH/kD is greater due to a
dϪ
O
␳
XZ ϭ –0.56) [20].
dϪ
The cross-interaction constant, ␳XZ, in the present
EtO9C SC6H4Z
work is determined by multiple regression of 20 k2
[kXZ in Eqs. (1) and (3)] values in Table I, Eq. (3). The
negative sign
HNdϩ Hdϩ
CH2
log (kXZ/kHH
)
C6H4X
ϭ Ϫ(0.65 Ϯ 0.03)␴ ϩ (1.49 Ϯ 0.03)␴
(3)
X
Z
Structure IV
Ϫ (0.47 Ϯ 0.14)␴ ␴
X
Z
greater degree of proton transfer for a stronger nucle-
of ␳ is an indication of the concerted process
XZ
ophile (␦␴ Ͻ 0) and nucleofuge (␦␴ Ͼ 0), which
X
Z
[17,18]. It is also to be noted that faster rates are ac-
companied by a larger magnitude of selectivity param-
eters, ␳X (␤X) and ␳Z (␤Z). The failure of the RSP also
supports the proposed concerted mechanism [17,18].
We therefore conclude that the aminolysis of thio-
phenyl derivatives, structure I (R ϭ EtO), is enforced
to proceed through a concerted mechanism due to de-
stabilization of the putative tetrahedral intermediate,
lead to a greater degree of bond cleavage (␦␳ Ͼ 0)
Z
␳
ϭ Ѩ␳Z /Ѩ␴ ϭ Ѩ␳X Ѩ␴ Ͻ 0
(4)
XZ
X
Z
and bond making (␦␳ Ͻ 0), respectively. The acti-
X
vation parameters in Table III are also in line with
those for a typical SN2-type concerted reaction.
Table II The Secondary-Kinetic Isotope Effects for the Reactions of O-Ethyl S-Aryl Thiocarbonates with Deuterated
X-Benzylamines in Acetonitrile at 45.0ЊCa
Ϫ
Ϫ
1
Ϫ
Ϫ1
X
Z
kH ϫ 103 (M 1s
)
kD ϫ 103 (M 1s
)
kH/kD
p-Me
p-Me
p-Me
p-Me
p-Cl
p-Cl
p-Cl
p-Cl
p-Me
H
p-Cl
p-Br
p-Me
H
16.8 Ϯ 0.08
26.9 Ϯ 0.1
65.2 Ϯ 0.5
71.8 Ϯ 0.6
9.70 Ϯ 0.02
15.5 Ϯ 0.06
31.6 Ϯ 0.2
36.1 Ϯ 0.3
12.3 Ϯ 0.06
19.1 Ϯ 0.09
39.8 Ϯ 0.2
41.9 Ϯ 0.4
7.52 Ϯ 0.05
10.4 Ϯ 0.07
18.9 Ϯ 0.1
20.2 Ϯ 0.2
1.37 Ϯ 0.01
1.41 Ϯ 0.01
1.63 Ϯ 0.02
1.71 Ϯ 0.02
1.28 Ϯ 0.01
1.49 Ϯ 0.01
1.67 Ϯ 0.02
1.78 Ϯ 0.03
p-Cl
p-Br
a Errors shown are standard deviations.

134
OH ET AL.
Table III Activation Parametersa for the Reactions of
O-Ethyl S-Aryl Thiocarbonates with X-Benzylamines in
Acetonitrile
MHz, CH3), 4.30 (2H, q, J ϭ 7.08 MHz, CH2),
7.39 ϳ 7.55 (5H, m, aromatic ring); and 13C NMR
(100.4 MHz, CDCl3), 168.7 (C"O) 136.2, 133.4,
127.5, 125.2 (aromatic), 64.1, 14.2.
Ϯ
Ϫ
1
Ϯ
Ϫ
Ϫ1
X
Z
⌬H /kcal mol
Ϫ⌬S /cal mol 1 K
p-OMe p-Me
p-OMe p-Br
16.3 Ϯ 0.1
14.4 Ϯ 0.1
12.7 Ϯ 0.1
12.5 Ϯ 0.1
15 Ϯ 1
18 Ϯ 1
28 Ϯ 1
26 Ϯ 1
O-Ethyl S-p-Methylphenyl-Thiocarbonate. The data
include liquid, IR(KBr), 2979 (C9H, CH2), 2939
(C9H, CH3), 1737 (C"O), 1596, 1488 (C"C, ar-
p-Cl
p-Cl
p-Me
p-Br
1
omatic), 1139, 1092 (C9O); H NMR (400 MHz,
a Calculated by the Eyring equation. Errors shown are standard
deviations.
CDCl3), 1.31 (3H, t, J ϭ 7.08 MHz, CH3), 2.37 (3H,
s, CH3), 4.29 (2H, q, J ϭ 7.08 MHz, CH2), 7.21–7.42
(4H, dd, J ϭ 7.81 MHz, aromatic ring); 13C NMR
(100.4 MHz, CDCl3), 169.9 (C"O), 139.8, 134.8,
129.9, 124.3 (aromatic), 63.9, 21.3, 14.3.
In summary, the reactions of O-ethyl S-(Z)aryl thio-
carbonates with (X) benzylamines in acetonitrile
proceed by a concerted displacement mechanism. This
conclusion is based on (i) the relatively small ␤X
(0.6 ϳ 0.8) and ␤Z (Ϫ0.5 ϳ Ϫ0.7) values, (ii) a neg-
O-Ethyl S-p-Chlorophenyl Thiocarbonate. Data
include liquid, IR(KBr), 2979 (C9H, CH2), 1723
(C"O), 1582, 1475 (C"C, aromatic), 1132, 1092
ative ␳ (Ϫ0.47) value, and (iii) the failure of the
XZ
1
(C9O); H NMR (400 MHz, CDCl3), 1.32 (3H, t,
RSP. The kinetic isotope effects, kH/kD Ͼ 1.0, suggest
that the TS has a four-center- type hydrogen-bonded
structure. It is notable that primary amines (benzylam-
ine) and acetonitrile as solvent destabilize the putative
tetrahedral intermediate, TϮ, so strongly as to enforce
a concerted mechanism, as found with secondary
amines (alicyclic amines in water), but not with terti-
ary amines (pyridines in water).
J ϭ 7.32 MHz, CH3), 4.30 (2H, q, J ϭ 7.08 MHz,
CH2), 7.47 ϳ 7.26 (4H, dd, 8.50 MHz, aromatic ring);
and 13C NMR (100.4 MHz, CDCl3), 168.9 (C"O),
136.0, 135.9, 129.3, 126.4 (aromatic), 64.3, 14.3.
O-Ethyl S-p-Bromophenyl Thiocarbonate. Data
include liquid, IR(KBr), 2979 (C9H, CH2), 1730
(C"O), 1575, 1474 (C"C, aromatic), 1152, 1092
1
(C9O); H NMR (400 MHz, CDCl3), 1.32 (3H, t,
J ϭ 7.08 MHz, CH3), 4.30 (2H, q, J ϭ 7.08 MHz,
CH2), 7.54 ϳ 7.38 (4H, dd, 8.50 MHz, aromatic ring);
and 13C NMR (100.4 MHz, CDCl3), 168.5 (C"O),
136.0, 132.1, 126.8, 124.0 (aromatic), 64.1, 14.1.
EXPERIMENTAL
Materials
Merck GR acetonitrile was used after three distilla-
tions. The benzylamine nucleophiles, Aldrich GR,
were used without further purification. Thiophenols
and ethyl chloroformate were Tokyo Kasei GR grade.
Kinetic Measurement
Rates were measured conductometrically at 45.0 Ϯ
0.05ЊC. The conductivity bridge used in this work was
a self-made computer automatic A/D converter.
Pseudo-first-order rate constants, kobs, were deter-
mined by the Guggenheim method [23] with a large
excess of benzylamine, [substrate] Х 0.001 M and
[benzylamine] Х 0.05 ϳ 0.1 M. Second-order rate
constants, k2, were obtained from the slope of a plot
of kobs vs. benzylamine with more than five concentra-
tions in more than three runs and were reproducible to
within Ϯ3%.
Preparations of O-Ethyl S-Aryl
Thiocarbonates [22]
Thiophenol derivatives and ethyl chloroformate were
dissolved in anhydrous ether and pyridine added care-
fully, keeping the temperature to 0 ϳ 5ЊC. Ice was
then added to the reaction mixture and the ether layer
was separated, dried on MgSO4, and distilled under
reduced pressure to remove the solvent. IR (Nicolet
1
5BX FT-IR) and H and 13C NMR (JEOL 400 MHz)
Product Analysis
data are cited next.
Substrate (0.05 mol) and benzylamine (0.5 mol) were
added to acetonitrile and reacted 45.0ЊC under the
same condition as the kinetic measurements. After
more than 15 half-lives, the solvent was removed un-
der reduced pressure and the product was separated by
O-Ethyl S-Phenyl Thiocarbonate. The data include
liquid, IR(KBr), 2979 (C9H, CH2), 1735 (C"O),
1592, 1478 (C9C, aromatic), 1135, 1087 (C9O);
1H NMR (400 MHz, CDCl3), 1.32 (3H, t, J ϭ 7.32

KINETICS AND MECHANISM OF THE AMINOLYSIS OF O-ETHYL S-ARYL
135
4. Castro, E. A.; Ureta, C. J Org Chem 1989, 54, 2153.
5. Castro, E. A.; Ureta, C. J Org Chem 1990, 55, 1676.
6. Cabrera, M.; Castro, E. A.; Salas, M.; Santos, J. G.;
Sepulveda, P. J Org Chem 1991, 56, 5324.
7. Castro, E. A.; Salas, M.; Santos, J. G. J Org Chem 1994,
59, 30.
8. Castro, E. A.; Ibanez, F.; Salas, M.; Santos, J. G. J Org
Chem 1991, 56, 4819.
9. Castro, E. A.; Ibanez, F.; Salas, M.; Santos, J. G.; Se-
pulveda, P. J Org Chem 1993, 58, 459.
10. Castro, E. A.; Cubillos, M.; Ibanez, F.; Moraga, I.; San-
tos, J. G. J Org Chem 1993, 58, 5400.
column chromatography (silica gel, 10% ethylacetate-
n-hexane). Analysis of the product gave the results
cited next.
CH3CH2OC("O)NHCH2C6H4 9OCH3. Data in-
clude mp45 ϳ 46ЊC, IR(KBr), 3315 (N9H), 2972
(C9H, benzyl), 2963 (C9H, CH2), 2946 (C9H,
CH3), 1683 (C"O), 1549 (C"C, aromatic), 1522
1
(N9H), 1260, 1038 (C9O); H NMR (400 MHz,
CDCl3), 1.99 ϳ 1.18 (3H, m, CH3 ), 3.73 (3H, s,
OCH3), 4.08 (2H, q, J ϭ 7.08, OCH2 ), 4.23 (2H, d,
Ϫ
11. Koh, H. J.; Han, K. L.; Lee, I. J Org Chem 1999, 64,
4783.
J ϭ 5.86, CH2 N), 5.55 (1H, s, NH), 7.18 ϳ 6.80
(4H, dd, J ϭ 8.30 MHz, aromatic ring); and 13C NMR
(100.4 MHz, CDCl3), 171.1 (C"O), 158.9, 156,9,
131.1, 128.9 (aromatic ring), 60.7, 55.2 (OCH3), 44.4
(CH2), 14.7 (CH3).
12. Lee, I. Chem Soc Rev 1990, 19, 317.
13. Lee, I. Adv Phys Org Chem 1992; 27, 57.
14. Lee, I. Chem Soc Rev 1995, 24, 223.
15. Pross, A. Adv Phys Org Chem 1977, 14, 69.
16. Buncel, E.; Wilson, H. J Chem Educ 1987, 64, 475.
17. Lee, I.; Lee, B. S.; Koh, H. J.; Chang, B. D. Bull Korean
Chem Soc 1995, 16, 277.
18. Lee, I.; Lee, H. W. Collect Czech Chem Commun, in
press.
19. Castro, E. A.; Pizarro, M. I.; Santos, J. G. J Org Chem
1996, 61, 5981.
The authors wish to acknowledge the financial support of
the Korea Research Foundation made in the program year
of 1998.
20. Oh, H. K.; Lee, J. Y.; Yun, J. H.; Park, Y. S.; Lee, I.
Int J Chem Kinet 1998, 30, 419.
BIBLIOGRAPHY
21. Hansch, C.; Leo, A.; Taft, R. W. Chem Rev 1991, 91,
165.
22. Al-Awadi, N.; Taylor, R. J Chem Soc Perkin Trans 2,
1986, 1581.
1. Jencks, W. P. Chem Rev 1985, 85, 511.
2. Williams, A. Chem Soc Rev 1994, 23, 93.
3. Gresser, M.; Jencks, W. P. J Am Chem Soc 1977, 99,
6963, 6970.
23. Guggenheim, E. A. Phil Mag 1926, 2, 538.