Skip to main content

Full text of "Chemical Sciences, Vol -92"

See other formats


Indian Academy of Sciences 
(Chemical Sciences) 



EDITORIAL BOARD 

C. N. R. Rao, Indian Institute of Science, Bangalore (Chairman and Editor of Publications, 
lASc) 

D. Balasubramanian, Centre for Cellular and Molecular Biology, Hyderabad 

A. Chakravorty, Indian Association for the Cultivation of Science, Calcutta 

B. M. Deb, Indian Institute of Technology, Bombay 
D. V. S. Jain, Panjab University, Chandigarh 

1. R. Kasturi, Indian Institute of Science, Bangalore 

V. Krishnan, Indian Institute of Science, Bangalore 

P. T. Manoharan, Indian Institute of Technology. Madras 

G. Mehta, University of Hyderabad, Hyderabad 

S. Rajappa, CIBA-GEIGY Research Centre, Bombay 

P. Ratnasamy, National Chemical Laboratory, Pune 

Sukh Dev, Malti-Chem Research Centre, Baroda 

B. Venkataraman, Tata Institute of Fundamental Research, Bombay 

Subscriptions 

For India, Pakistan, Sri Lanka, Nepal, Bangladesh and Burma, contact: 
Editorial Office 

For other countries, contact: 

M/s J C Baltzer A. G. 
Scientific Publishing Company 
Wettsteinplatz 10 
CH-4058 Basel 
Switzerland 

Telephone: 61-268925. Telex: 63475 



Editorial Office: 

Indian Academy of Sciences 
P.B. No. 8005 
Bangalore 560080 
India 

Notes on the preparation of papers 

Authors wishing to have papers published in the Typescript: 
Proceedings should send them to Editorial Office. 

Papers should be typed double spaced with ample 
Three copies of the paper must be submitted. margin on all sides on white bond paper 

280x215 mm. This also applies to the abstract, 

The papers must normally present results of original tables, figure captions and the list of references 
work. Critical reviews of important fields will also be which are to be typed on separate sheets, 
considered for publication. Submission of the script T|H 
will be held to imply that it has not been previously P " ge: 



Absorption and emission spectra of isomeric tolunitriles 

A MAITI, S K SARKAR and G S KASTHA* 

Optics Department, Indian Association for the Cultivation of Science, Jadavpur, C; 
700032, India 

MS received 9 May 1983; revised 13 July 1983 

Abstract. Absorption and emission characteristics of o, m- and p-tolunitriles in pol 
non-polar solvents under different conditions have been investigated in detail. Solvatocl 
shifts of band origin of these molecules in non-polar solvents show that their dipolemc 
in the first excited singlet state are almost the same while its value in the second excited sii 
larger in the meta- than in the para-isomer. Vibronic analyses of the low tempei 
absorption, fluorescence and phosphorescence spectra of all the three molecules have pr 
evidence that these molecules are slightly distorted in the first excited singlet state whi 
distortion in the phosphorescence emitting triplet state is larger. The data on fluorescer 
phosphorescence quantum yield and phosphorescence lifetime of the tolunitril 
reasonably interpreted as showing that in these molecules, particularly m- and p- tolur 
the internal conversion rate from the first excited singlet to the ground state is probabl; 
and that the charge transfer character of the triplet state in the p-isomer is larger than 
the meta. 

Keywords. Isomeric tolunitriles; dipole moment; molecular geometry; CT-character; ra 
and non-radiative rates. 



1. Introduction 

Benzonitrile which belongs to C 2t) -point group exhibits two systems of absor 
bands of it, n* character in the near uv region designated as 1 A 1 1 A l and ' 1 JB 2 - 
and correspond to 1 L a 1 A and 1 L b *A transitions respectively under Platt not. 
Vapour phase absorption measurements (Hirt and Howe 1948) have led t< 
conclusion that the first excited singlet it, n* state of the molecule possesses 
intramolecular charge-transfer (CT) and local excitation character. According to I 
and Knight (1970) the phenyl nucleus acts as an electron donor in the cr-pr< 
Furthermore, LeBel and Laposa (1972) have shown that the phosphorescence em: 
state of benzonitrile is a 3 X 1 (n,it*) level and the molecule in the first excited si 
state is a nearly regular but slightly expanded hexagon, while in the first excited t 
the molecule is planar but a non-regular hexagon. In connection with the EPR stu 
the triplet states of fluorobenzonitriles Wagner and May (1976) observed that th< 
excited triplet state of benzonitrile is a 1,4-diradical, often referred to as the 'quir 
form of the triplet. 

Among the substituted benzonitriles, the phosphorescence of three iso] 
tolunitriles were studied by Takei and Kanda (1962) who concluded that the emi 
characteristics of the three molecules should be attributed to the electronic tran: 



2 A Maiti, S K Sarkar and G S Kastha 

based on resonance between the nitrile group and the phenyl ring. Later, Lui and 
McGlynn (1975a, b) investigated the absorption and emission characteristics of three 
isomeric cyanoanisoles and fluorbenzonitriles and observed certain similarities in the 
variations of spectral characteristics among the corresponding members of the two 
triads. For example, in each triad, the p- isomer has the smallest l L a 1 L b energy gap, 
highest 1 L b *- 1 A transition energy with the smallest /-value and the lowest 1 L a +- l A 
transition frequency having the largest /-value. These authors, on the basis of their 
experimental results, concluded that in all the substituted benzonitriles, the phos- 
phorescence emitting triplet states have varying amount of ex-character. However, they 
made no attempt to find out if the geometries of these molecules are affected in the 
excited singlet and triplet states. 

It is known that the methyl group is a weaker electron donor than either the fluorine 
atom or the methoxy group and therefore, it would be interesting to investigate if the 
spectral characteristics of the three isomeric tolunitriles show similar variations as 
found in the two above mentioned triads. It would also be in order to find out, at least 
qualitatively, if the excited states of these molecules possess any cr-character, whether 
the geometries of these molecules in the excited states are altered and how these excited 
state properties are related to the phosphorescence characteristics of the molecules. 
With these objectives, the absorption, fluorescence and phosphorescence of the three 
isomeric tolunitriles in polar and non-polar solvents have been investigated. The results 
obtained and discussion of these results form the subject matter of the present 
communication. 



2. Experimental 

Highly pure samples of m- and p-tolunitriles were obtained from Koch-Light 
laboratories (England) and o-tolunitrile was procured from Fluka (Switzerland). The 
samples were distilled several times till no impurity could be detected by the GLC- 
method. The solvents ethanol, n-hexane and methylcyclohexane (MCH) were of 
spectrograde quality from E. Merck (W. Germany) and they were carefully distilled 
under reduced pressure before being used in the investigation. De-oxygenated solutions 
of the tolunitriles of ca. 10~ 4 M concentration were prepared by the usual freeze and 
thaw technique and their fluorescence and phosphorescence spectra were recorded on a 
Perkin-Elmer MPF-44A spectrofluorimeter equipped with a compensated spectral 
unit. The phosphorescence life-time (T P ) was measured following the decay of 
phosphorescence signal on the chart recorder having a response of 0.3 sec. The 
fluorescence and phosphorescence quantum yields (<D f and O p respectively) in ethanol 
glass (EtOH) were determined by the standard method using benzonitrile as reference. 
In the case of MCH glass the ratio of the yields (<D P /<D f ) was computed from the ratio of 
the areas under phosphorescence and fluorescence spectra respectively. The errors in 



3.1 Absorption spectra 

Like benzonitrile, all the tolunitriles show two systems of absorption bands in the near 
uv region. The first system of bands at the lower energy has a low /-value and 
corresponds to l L b <- M transition while the second system with higher energy and 
moderately large /-value is attributed to l L a +- 1 A transition. If the CH 3 -group is 
treated as a single mass point, then in the reduced symmetry C s of o- and m-tolunitriles 
both these transitions are 1 A' 1 A' while in p-tolunitrile (C 2t ,-point group) the 
designations of the transitions are the same as in benzonitrile. For each isomer the 
vapour phase absorption spectra undergo bathochromic shifts in solutions and these 
shifts for the 1 L a -band are larger than those for the 1 L fc -band. The absorption spectra of 
the compounds recorded in rigid EtOH and MCH glasses at 77 K show slight shifts with 
respect to those recorded at room temperature but the vibrational structures in the 
former are more pronounced. The relevant data on the absorption of the tolunitriles are 
presented in table 1 while their vibrational assignments are shown in tables 3a, b and c. 
For comparison the vibrational frequencies as determined by Padhye and 
Varadarajan (1962) from vapour phase absorption measurements on the tolunitriles 
are included in the relevant tables. 

Analysis presented in these tables shows that in the absorption spectra of each of the 
isomeric tolunitrile molecules in EtOH and MCH glasses, at 77 K there are, besides a few 
fundamental excited state frequencies, progressions involving the frequencies 730 and 
757 cm ~ * in o-tolunitrile, 958 and 996 cm~ l in m-tolunitrile and 778 and 789 cm" * in 
p-tolunitrile. These frequencies most probably correspond to the ground state ring 
vibration frequencies 715, 995 and 815 cm" 1 in o-, m- and p-tolunitriles respectively. 

Table 1. Absorption spectral data of benzonitrile and isomeric tolunitriles at room 
temperature. 



^L a band system 


l L b band 


system 

v tif \ v (i r \ 


Molecule 


Phase 


v 00 (cm~ 1 ) 


/ 


voo(cm-) 


v \ *-'a) V \ '-'b! 

f (cm" 1 ) 


Benzonitrile 


vapour 


44740 





36500 


8240 




n-hex 


43465 


0-22 


36221 


0-012 




MCH 


43277 


0-22 


36090 


0-012 




EtOH 


43277 


0-22 


36090 


0-012 


o-tolunitrile* 


vapour 


43855 





35770 


8085 




n-hex 


44234 


0-22 


35388 


0-019 




MCH 


43655 


0-24 


35263 


0-018 




EtOH 


42540 


0-25 


35139 


0-030 


m-tolunitrile 


vapour 


44430 





35806 


8624 




n-hex 


43183 


0-24 


35451 


0-021 - 




MCH 


43090 


0-22 


35356 


0-018 




EtOH 


42722 


0-27 


35263 


. 0-025 


p-tolunitrile 


vapour 


43185 





36204 


7981 




n-hex 


42181 


0-30 


35961 


0-010 




MCH 


42181 


0-30 


35864 


0-008 



Molecule 



>L a -band 
system 



1 Lj,-band 

system 



Dipole moment D 



Solvent 



Av(cm 



a 3 in Ground 

(AU 3 ) state l L b l L a 



Benzonitrile 


n-hex 


1275 


279 






MCH 


1463 


410 


20 4-10 4-81 5-83 




EtOH 


1463 


410 




y.l-^1 average 




l-40a 3 - 


0-35a 3 - 








10-80 


0-71 




o-tolunitrile 


n-hex 




382 






MCH 




507 


25 3-95 5-07 




EtOH 




631 




\t\-\i\ average 






0-45a 3 - 










1-12 




m-tolunitrile 


n-hex 


1247 


355 






MCH 


1343 


450 


25 4-25 5-21 6-31 




EtOH 


1708 


543 




-\il average 




l-32a 3 - 


0-41a 3 - 








11-35 


1-18 




p-tolunitrile 


n-hex 


1004 


243 






MCH 


1004 


340 


25 4-40 5-23 5-43 




EtOH 


1181 


310 




-\i\ average 




l-03a 3 - 


0-30a 3 - 








15-65 


0-53 





* Solvent shift Av = v 00 (gas)-v 00 (solution) cm" 



The frequencies 1214 and 1222 cm r in o-tolunitrile, 1263 and 1289 cm * in the m- and 
1176 and 1198cm" 1 in the p-isomer are believed to represent the valance v(C-CN) 
vibration frequencies in the excited state of these molecules. It should be noted that 
Padhye and Varadarajan (1962) observed this frequency at 1199, 1251 and 1175 cm' 1 
in the vapour phase absorption spectra of the o-, m- and p-tolunitriles respectively. In 
the above the two entries for each compound refer to data from the spectra in the two 
solvents. 

3.2 Fluorescence emission spectra 

The tolunitrile molecules exhibit well-defined band origin and structured fluorescence 
in the rigid MCH and EtOH glasses at 77 K. Since all of them are very similar, the spectra 
of only one, viz o-tolunitrile are reproduced in figure 1. For comparison the 
corresponding l L b *- l A absorption spectra recorded at the same temperature are also 
shown in the same figure and the fair mirror image symmetry between them is evident. 
Accordingly, the fluorescence is attributed to the reverse l L b -* * A transition. From the 
vibrational analyses of these spectra (only one is presented in table 4) it is found that, as 
in absorption a few ground state frequencies in addition to a short progression 
involving ring vibration is observed in each of the tolunitriles. These frequencies are 760 



Table 3. Vibrational analysis of the absorption spectra at 77 K. 



MCH-glass 
vfcm" 1 ) Av(cm~ l ) assignment 



v(cm" 



EtOH-glass 

A v(cm ~ l ) assignment 



(a) Orthotolunitrile 


35263 





0,0 


35139 





0,0 


35767 


504 


+ 504 


35576 


437 


+ 437 


35993 


730 


+ 730 


35896 


757 


+ 757 


36485 


1222 


0+1222 


36353 


1214 


+ 1214 


36760 


1497 


+ 2x730 


36700 


1561 


+ 2x757 


37298 


2035 


0+730+1222 


37133 


1994 


+ 757 + 1214 


37442 


2179 


+ 3x730 


37539 


2400 


+ 3x757 


38083 


2820 


+ 4x730 


38156 


3017 


+ 4x757 


38376 


3113 


5x730 


38748 


3609 


+ 5x757 






(b) 


Metatolunitrile 






35356 





0,0 


35263 





0,0 


35800 


444 


+ 444 


35640 


377 


+ 377 


36090 


734 


+ 734 


36221 


958 


+ 958 


36352 


996 


+ 996 


36552 


1289 


0+1289 


36619 


1263 


0+1263 


37232 


1969 


+ 2x958 


36991 


1635 


+ 734 + 996 


38193 


2930 


+ 3x958 


37302 


1946 


+ 2x996 








37582 


2226 


+ 996+1263 








38266 


2910 


+ 3 x 996 












(c) 


Paratolunitrite 






35864 





0,0 


35894 





0,0 


36485 


621 


+ 621 


36419 


523 


+ 523 


36653 


789 


+ 789 


36684 


778 


+ 778 


37040 


1176 


+ 1176 


37094 


1198 


0+"ll98 


37407 


1543 


+ 2x789 


37442 


1548 


+ 2x778 


37796 


1932 


+ 789 + 1176 


37864 


1970 


0+778 + 1198 


38229 


2365 


+ 3x789 


38229 


2335 


+ 3x778 


38598 


2734 


+ 2x789+1176 38598 


2704 


+ 2x778 + 1198 


38957 


3111 


+ 4x789 


38974 


3080 


+ 4x778 


39749 


3885 


+ 5x789 


39749 


3855 


+ 5x778 



Vapour phase absorption v 00 for a. 35770cm l b. 35806cm l and c. 36204cm" 1 . The 
excited state frequencies for a. 669, 961 and 1 199 cm~ l . b. 672, 970 and 1251 cm" 1 c. 507, 761 
and 11 75 cm" 1 (Padhye and Varadarajan). 



1171 and 1178 cm x and 1216 cm 1 in the o-, m- and p-tohmitrile respectively. These 
frequencies in the Raman spectra as reported by Takei and Kanda (1962) are 
respectively at 1209, 1245 and 1194 cm~ 1 for o-, m- and p-tolunitriles. The two entries 
have the same connotation as given in the preceding paragraph. 

3.3 Phosphorescence emission 

Like fluorescence the phosphorescence spectra of the tolunitriles in MCH and EtOH 
glasses show vibrational structures. These spectra, one of which is shown in figure 2, are 




270 290 310 

WAVELENGTH (rim) 



330 



Figure 1. Fluorescence and absorption spectra for o-tolunitrile in EtOH ( 
( )at77K. 



-),andMCH 




360 



390 



480 



420 450 

Wavelength (nm) 

Figure 2. Phosphorescence emission spectra (P) of p-tolunitrile in EtOH ( - 
( )at77K. 

Table 4. Vibrational analysis of fluorescence spectra at 77 K. 



-)andMCH 



v(cm" 



EtOH-glass 



assignment 



MCH-glass 

v(cm ~ 1 ) Av(cm " * ) assignment 



Orthotolunitrile 


35170 





0,0 


35263 





0,0 


34757 


413 


0-413 


34848 


415 


0-415 


34410 


760 


0-760 


34488 


775 


0-775 


33961 


1209 


0-1209 


34090 


1173 


0-1173 


33618 


1552 


0-2x760 


33703 


1560 


0-2x775 


32872 


2298 


0-3x760 


33323 


1940 


0-775-1173 



c s = o 



Molecule 
(medium 77 K) 


*, 


*. 


<t> P /<t>F 


(sec) 


(sec *) 


(sec- 1 ) 


(sec" 1 ) 


(sec' 1 ) 


(cm- 1 ) 


o-Tolunitrile 




















EtOH 


0-36 


0-24 


0-67 


3-40 


4-0 


0-18 


0-24 


0-29 


8505 


MCH 






0-51 


3-90 










8736 


m-tolunitrile 




















EtOH 


0-51 


0-23 


0-45 


4-60 


1-4 


0-10 


0-086 


0-22 


8795 


MCH 






0-80 


4-40 










8961 


p-tolunitrile 




















EtOH 


0-36 


0-33 


0-92 


3-80 


2-0 


0-14 


0-099 


0-26 


9332 


MCH 






HO 


3-40 










9344 


d-fluorobenzonitrile 




















EtOH 


0-51 


0-093 


0-18 


2-43 


1-9 


0-077 


0-36 


0-41 


8700 


m-fluorobenzonitrile 




















EtOH 


0-54 


0-082 


0-15 


2-60 


1-9 


0-068 


0-35 


0-38 


8900 


p-fluorobenzonitrile 




















EtOH 


0-24 


0-23 


0-96 


2-05 


2-4 


0-15 


0-71 


0-49 


9300 


o-cyanoanisole 




















EtOH 


0-26 


0-11 


0-42 


1-40 


14-0 


0-11 


2-20 


0-71 


7310 


m-cyanoanisole 




















EtOH 


0-21 


0-094 


0-45 


1-80 


15-0 


0-066 


1-90 


0-86 


8070 


p-cyanoanisole 




















EtOH 


0-22 


0-24 


1-00 


1-60 


10-0 


0-18 


3-10 


0-63 


9040 



The vibrational assignment of the phosphorescence bands and the second-order 
magnitude of the phosphorescence lifetime indicate the n, n* nature of the phosphores- 
cence emitting state in these molecules. Following LeBel and Laposa's (1972) 
assignment of the phosphorescence emitting state in benzonitrile, which is 3 A^ (n, 7t*), 
the corresponding state in each of o- and m-tolunitriles may be designated 3 /4' (n, n*). 
The data on fluorescence and phosphorescence characteristics are collected in table 5 
and for comparison those for cyanoanisoles and fluorobenzonitriles (Lui and McGlynn 
1975a, b), are included in this table. 

4. Discussion 

From the data on the absorption spectra of the three tolunitriles (table 1), it is seen that 
the p-isomer shows spectral features which are different from those of the other two. 
For example, the ^-band in p-tolunitrile has the highest frequency and the lowest 
/-value while the energy of the l L a band is the lowest but its j^-value is the largest. 
Moreover, the l L a l L b energy gap is smallest for this molecule. These features are 
similar to those observed in the isomeric cyanoanisoles and fluorobenzonitriles by Lui 
and McGlynn (1975a, b). Following Suzuki (1967) the higher energy of the 1 L b <- 1 A 
transition in p-tolunitrile as compared to that of the o-isomer may be understood in 



tne CT-siaie wun me locauy cxcueu suues 01 oenzcne wm lower me energy or tne nrst 
transition of the o-molecule more than that of the p-isomer. 

4.1 Estimate of ci-character in the excited singlet states 

To estimate the relative cr-character in the first and second excited n, n* singlet states, 
denoted respectively by Si and S 2 , of the isomeric tolunitriles, the data on the 
solvatochromic shifts of band origin, summarized from table 1, are presented in table 2 
and analysed. The solvent-shift data in the non-polar n-hexane and MCH solvents only 
are considered and those in EtOH solvent are excluded because of the known 
complications arising with the hydrogen bonding ethanol. The shift of the positions of 
the origin of l L a - and ^-band systems in going from vapour phase to the non -polar 
solvents may be represented by the equation, 

. gas Solution -.. 10 9 /* 5031 , ., , fl 2 1 

Av = v 00 v 00 = 



va" a" " 2n" + i 

where the first term on the right side expression is the dispersion term of Bayliss (1950) 
and the second term represents the difference in the electrostatic energies of the excited 
and ground state dipolemoments embedded in a continuum of dielectric constant n 2 
(Mataga and Kubota 1970). In this equation, v and Av are in cm" l , a is the cavity radius 
in AU, n e and \i g are the dipolemoments (in Debye unit) of the excited and ground states 
respectively, n is the refractive index of the solvent and /-represents the oscillator 
strength of the transition considered. From the data on the ^-band of benzonitrile 
presented in table 2, the \i\ values calculated from the above relation are 0-30 a 3 

0-71 in rc-hexane and 0-40 a 3 0-71 in MCH with a mean value given by Atf /^ = 
0-35 a 3 -0-71. Similarly for the ^-band, tf-fJ-g = 1-36 a 3 - 10-78 in n-hexane and 1-43 a 3 

10-83 in MCH with the average 1-40 a 3 10-81. In both the cases the ^ n% value 
computed for n-hexane and MCH solvents are about 15 % of which, taking into account 
the uncertainties in the solvent-shift values, may be considered fairly satisfactory. The 
value of the dipole moment of benzonitrile in the ground state (S ) is 4- ID (Brown 
1959) and if a value of 20 A 3 is assumed for a 3 , which may not be unreasonable, the 
dipole moments fi e in the i L b (S l ) and l L a (S 2 ) states work out as 4-81 and 5-83D 
respectively. The former may be compared with 4-45D in the first excited singlet (S) 
state estimated by Hirst and Howe (1948). These dipole moment values show that the 
ci-character in the S t state is small and is moderate in the S 2 -state. It is noted that 
though the actual n e value is dependent on the assumed a 3 value nonetheless, for 
reasonable a 3 values \i e in the S 2 -state is always greater than that in the S^ -state. 

The expressions for nj - \L\ relating to the 1 L a (S 2 ) and * L b (S t ) states of the three 
isomeric tolunitriles derived in the same way as in benzonitrile are shown in table 2 
which also contain the /^-values for the two states. For p-tolunitrile the reported value 
of n g = 4-4 D (Brown 1959) is used and for the other two molecules, for which 
experimental data are not available in the published literature, the respective ^ 9 -value 
has been derived by the vector-composition method. The computed values are 3-95 and 
4-25 D for the o- and m-tolunitrile respectively. The ^-values in table 2 have been 

Calculated with a 3 = 25 A 3 which is a little lareer than that nspri with hp.nTrvnitrilp, 



2 OLd 2 *" U A w A W ** v O 

ci-character than the p-isomer which in turn seems to possess nearly the same 
ex-character as benzonitrile. The /vvalue of o-tolunitrile in the 5 2 -state could not be 
calculated because it has not been possible to determine the position of band origin 
owing to the broadness of its ^-band even in the non-polar solvents. 

4.2 Excited state molecular geometries 

From the vibronic analysis of the absorption data in EtOH and MCH glasses for the 
three isomeric tolunitriles presented in table 3a, b and c it is seen that in all cases 
moderately long progressions of ring vibration frequencies are present. In p-tolunitrile 
vibronic band with five quantum of excitation of ring frequency is observed while in the 
o- and m-compounds respectively four and three quantum of excitations of this 
vibrational mode are evident. Moreover, the v(C-CN) vibrational frequency and its 
combination with the ring vibration frequency are also found. Similarly in the 
fluorescence spectra of these compounds (table 4) the ring vibration frequency seems to 
form the more prominent progression. These two observations are complimentary and 
they point to the fact that in the fluorescent ^-singlet state each of the tolunitrile 
molecules suffers small ring distortions in comparison with its geometry in the ground 
state (S ). These observations and the conclusion derived therefrom is broadly the same 
as those made by LeBel and Laposa (1972) from the analyses of the fluorescence spectra 
of benzonitrile and its various deuterated derivatives. The coupling of v(C-CN) 
frequency probably indicates that the C-CN bond is also slightly affected in the 
Si -excited state. 

The characteristics of the phosphorescence emissions of the isomeric tolunitriles in 
EtOH and MCH glasses have been presented in a previous section. From the vibronic 
data derived from these spectra, it is found that the phosphorescence spectrum of each 
of these molecules shows progression of a vibrational frequency (ca. 1600cm" 1 ) 
corresponding to the v(C=C) of benzene and combinations of this frequency with ring 
vibration and valence v(C-CN) frequencies. These results are again similar to that 
reported by LeBel and Laposa (1972) from the analyses of the phosphorescence spectra 
of benzonitrile and the different deutero substituted benzonitriles. Hence, their 
conclusion that the benzonitriles in the phosphorescence triplet state (7\ ) are non- 
regular planar hexagons is also applicable to the tolunitriles. However, no definite 
suggestion about the shape of the molecules in the triplet state (7\ ) could be made. 

4.3 Interpretation of the variation of phosphorescence quantum yields 

From the results of the present investigation it is found that the isomeric tolunitriles 
exhibit variations in their luminescence properties similar to those observed in isomeric 
cyanoanisoles and fluorobenzonitriles reported earlier (Lui and McGlynn 1975a, b). A 
careful examination of the data on the luminescence properties of the three substituted 
benzonitriles presented in table 5 reveals the following trends. 

(i) The <J> p /O f -value is the highest in the para compound which also has the largest 
O p -value. (ii) The phosphorescence lifetime (T P ) is largest for the m-isomer for which 



10 A Maiti, S K Sarkar and G S Kastha 

separation, (iv) The O p - and r p - values for any member of the tolunitriles are larger than 
those of the corresponding member of the other two triads. 

An attempt to rationalize the observed variations in the <l> p -values of the isomers of 
the tolunitrile molecules has been made following the observations of Lui and 
McGlynn (1975a) in cyanoanisoles. For this purpose the values of intersystem crossing 
rate (K lsc ) and phosphorescence radiative rate (K p ) have been derived for the two ex- 
treme cases, in the first of which internal conversion rate (K s ) for the process SpMx->S 
is assumed to be zero and the second for which nonradiative rate (K T ) from the phos- 
phorescent T 1 ! -state, i.e. T^-vvw.^ is neglected as has been done by Lui and McGlynn 
(1975a, b). These are shown in table 5. From the -K ISC , K p and O p values shown in 
table 5 it is found that the variation of <1> P among the isomers of any of these triads 
roughly follows the variations of their K ISC and K p - values calculated on either of these 
two assumptions. 

Actually neither K s nor K T is entirely negligible and the choice of one of the extremes 
requires justification. It is seen from table 5 that for any of the tolunitriles its 
phosphorescence lifetime in EtOH or MCH medium is not much different from one 
another, which suggests that the surrounding environment does not have much 
influence on the nonradiative rate constant K T and that it is essentially a molecular 
property. It has been found from the previous discussions that the tolunitrile molecules 
are slightly distorted in the first excited state (5j ) while such distortions are greater in 
the triplet state (Tj ). Moreover, the Si S energy difference is much larger than the 
7\ S separation. Consequently, the internal conversion rate K s is expected to be 
small and the nonradiative rate K T to be non-negligible. These arguments seem to 
favour the first choice for which K s is assumed to be zero. However, this choice is not 
entirely satisfactory. For example, in o-tolunitrile though K lsc and K p are both larger 
than those of p-tolunitrile, the <I> P of the former is less than that of the latter. This 
discrepancy is also seen in the cyanoanisoles but not in the fluorobenzonitriles. Most 
probably the steric hindrance between the substituent and the nitrile group in both o- 
tolunitrile and o-cyanoanisole produces the observed discrepancy. 

It is also seem from table 5 that for the limit K s = 0, the phosphorescence radiative 
rate K p for the m-isomer is smaller than that for the p-compound in any of the triads. 
This is consistent with the lower O^-value for the former than that of the latter. It has 
been shown that in the tolunitriles the cr-character of the S t -state in the m- and 
p-compounds is probably the same while that in the S r 2 -state it is greater in m- than in 
p-tolunitrile. It is known that the radiative rate (K p ) from the lowest triplet state (7i ) 
depends on the amount of singlet character induced in it due to coupling with the 
various excited singlets states and the coupling is controlled by the spin orbit matrix 
elements between the various excited singlets and the triplet in question. Hence the 
higher cr-character of 5 2 -state in m-tolunitrile would give rise to a larger spin orbit 
matrix in this molecule than in the para, unless the combining triplet (7i ) state in the 
latter has much more cr-character than in the former. This conclusion is in agreement 
with the suggestion bv Lui and McGlvnn (1975a) in exolainine the ohosohorescence 



Brown T C 1959 J. Am. Chem. Soc. 81 3232 

Hirt R C and Howe J F 1948 J. Chem. Phys. 16 480 

LeBel G L and Laposa J D 1972 J. Mol. Spectrosc. 41 249 

Lui Y H and McGlynn S P 1975a J. Mol. Spectrosc. 55 163 

Lui Y H and McGlynn S P 1975b J. Lumin. 9 449 

Mataga N and Kubota T 1970 Molecular Interactions and Electronic Spectra (New York: Marcel Dckker 

Inc.) p. 384 

Padhye M R and Varadarajan T S 1962 J. Sci. Ind. Res. B21 241 
Suzuki H 1967 Electronic Absorption Spectra and Geometry of Organic Molecules (New York, London: 

Academic Press) 498 

Takei K and Kanda Y 1962 Spectrochim. Acta 18 1201 
Wagner P J and May M J 1976 Chem. Phys. Letts. 39 350 



Spectrophotometric determination of basicities of substituted 
acetyibiphenyls and biphenyi carboxylic acids 

P ANANTHAKRISHNA NADAR* and N KANNAN 

Department of Chemistry, VHNSN College, Virudhunagar 626002, India 

* Chemistry Department, Anna University, Guindy Campus, Madras 600025, India 

MS received 22 November 1982; revised 24 February 1983 

Abstract. The basicities of several 2'-, 3'-, and 4'-substituted 4-acetylbiphenyls and biphenyl- 
4-carboxylic acids have been determined spectrophotometrically in sulphuric acid media at 
30 U C. The pK BH+ of 3'- and 4'-substituted compounds are correlated by the Hammett 
equation. The 4'-methoxy group deviates considerably in the Hammett plot. This is attributed 
to its conjugative interaction with the carbonyl or carboxyl group aided by protonation. Good 
correlation exists between pK BH + and CT + . The basicities of 2'-substituted 4-acetylbiphenyls 
and biphenyl-4-carboxylic acids reaffirm the existence of jr-electron steric effect of 2'- 
substituents. 

Keywords. Basicities; acetyibiphenyls; biphenyi carboxylic acids; Hammett equation; n- 
electron steric effect. 



1. Introduction 

The most important property of sulphuric acid-water mixtures from the point of view 
of their usefulness as reaction media is their acidity, measured in terms of their acidity 
function. The investigation of reactions in strongly acid media by Hammett (1932), 
Stewart and Yates (1958), Bunnett and Olsen (1965, 1966) and Arnett and Anderson 
(1963) has in recent years generated considerable interest. The importance of the work 
is brought out by Liler (1971) and Paul (1957). The measurement of the extent of 
protonation in these media, in conjunction with the acidity functions that the various 
types of bases obey, results in a large number of pK values whose correlation often leads 
to new insight into the analysis of the effect of structure on reactivity. 

Sulphuric acid media offer an extremely wide and continuous range of acidity, 
joining on to the dilute aqueous range in which a large number of normally neutral 
compounds are protonated. The basicity of a large number of 2'-, 3'- and 4'-substituted 
4-acetylbiphenyls and biphenyl-4-carboxylic acids was determined by the Spectro- 
photometric technique to test the applicability of Hammett equation to the data on 3'- 
and 4'-substituted compounds. 

2. Results and discussion 

The results are given in tables 2-5. While determining the pK BH + of a base, it is 
important to choose the right acidity function such that pK BH+ is thermodynamically 



To whom all correspondence should be addressed. 



Substituent 


m.p. () 


Formula 


Observed( %) 
C H 


Required(%) 
C H 


Substituted 4-acetylbiphenyls 


H 


120-121 (121)" 


C 14 H 12 


87-8 


6-2 


87-7 


6-1 


3'-F 


91-92 (90-5)" 


C 14 H U FO 


78-3 


5-3 


78-5 


5-1 


3'-Cl 


58-59 (57-5-58-5)' 


C 14 H U C1O 


72-8 


4-9 


72-7 


4-8 


3'-Br 


45^6 (45-46)" 


C 14 H n BrO 


61-4 


4-1 


61-1 


4-0 


3'-NO 2 


110-111 (110-111)" 


C 14 H M N0 3 


69-7 


4-8 


69-7 


4-6 


3'-CH 3 


78-79 


C 15 H 14 


85-8 


6-8 


85-7 


6-7 


3'-OCH 3 


89-90 


Ci5Hi 4 O 2 


79-8 


6-3 


79-7 


6-2 


4'-F 


105-106 (105-106)" 


C 14 H U FO 


78-6 


5-3 


78-5 


5-1 


4'-Cl 


103-104 (103-104)" 


C, 4 H U CIO 


72-8 


4-7 


72-7 


4-8 


4'-Br 


130-131 (131) d 


Cj 4 H n BrO 


61-0 


4-1 


61-1 


4-0 


4'-NO 2 


153-154 (152-153) e 


C 14 H U N0 3 


69-9 


4-5 


69-7 


4-6 


4'-CH 3 


122-123 (122K 


C 15 H 14 


85-8 


6-9 


85-7 


6-7 


4'-OCH 3 


155-156 (153-154) 9 


CijHi 4 O 2 


79-6 


6-3 


79-7 


6-2 


2'-F 


85-86 (86-87)" 


C 14 H U FO 


78-4 


5-3 


78-5 


5-1 


2'-Cl 


56-57 (54-56)" 


C 14 H U C1O 


72-8 


4-9 


72-7 


4-8 


2'-Br 


81-82 (81-82)" 


C, 4 H u BrO 


61-3 


4-2 


61-1 


4-0 


2'-N0 2 


110-111 (110) e 


C 14 H U N0 3 


69-9 


4-8 


69-7 


4-6 


2'-CH 3 


52-53 


C 15 H 14 


85-6 


6-8 


85-7 


6-7 


2'-OCH 3 


63-64 


CisH 14 O 2 


79-6 


6-1 


79-7 


6-2 


Substituted biphenyl-4-carboxylic acids' 1 


H 


225-226 (225-8) 


Ci3H 10 O 2 


78-9 


5-3 


78-8 


5-1 


3'-F 


241-242 (240-241-5) 


C 13 H 9 F0 2 


75-3 


4-4 


75-4 


4-3 


3'-Cl 


248-249 (249-250) 


C 13 H 9 CI0 2 


67-1 


4-0 


67-0 


3-9 


3'-Br 


253-254 (252-9-254-4) 


C 13 H 9 BrO 2 


56-2 


3-5 


56-3 


3-3 


3'-NO 2 


314-315 (313-315) 


C 13 H 9 N0 4 


64-3 


3-8 


64-2 


3-7 


3'-CH 3 


205-206 (206-207) 


Cj 4 H 12 O 2 


79-4 


5-9 


79-3 


5-7 


3'-OCH 3 


197-198 (197-198) 


C 14 H ]2 O 3 


73-9 


5-7 


73-7 


5-3 


4'-F 


237-238 (236-238) 


C I3 H 9 F0 2 


75-6 


4-2 


75-4 


4-3 


4'-Cl 


293-294 (290-293) 


C 13 H 9 C10 2 


67-1 


4-0 


67-0 


3-9 


4'-Br 


304-305 (303-305) 


C 13 H 9 BrO 2 


56-1 


3-5 


56-3 


3-3 


4'-N0 2 


338-340 (336-338) 


C 13 H 9 NO 4 


64-3 


3-8 


64-2 


3-7 


4'-CHj 


242-243 (243-245) 


C 14 H 12 O 2 


79-4 


5-8 


79-3 


5-7 


4'-OCH 3 


248-249 (247-248) 


C 14 Hi 2 O 3 


73-9 


5-4 


73-7 


5-3 


2'-F 


231-232 (232-233) 


C 13 H 9 F0 2 


75-6 


4-2 


75-4 


4-3 


2'-Cl 


251-252 (251-5-252-5) 


C 13 H 9 C1O 2 


69-2 


3-9 


69-0 


3-9 


2'-Br 


242-243 (242) 


\~>1 jiigOrOj 


56-5 


3-4 


56-3 


3-3 


2'-NO 2 


253-254 (253-5-254-4) 


C 13 H 9 N0 4 


64-4 


3-6 


64-2 


3-7 


2'-CH 3 


205-206 (206-207) 


C 13 H 12 O 2 


79-5 


6-0 


79-3 


5-7 


2'-OCH 3 


197-198 (197-198) 


C 14 H 12 O 3 


74-0 


5-6 


73-7 


5-3 



Values in parentheses are literature values 

" Long and Henze (1941); * Byron et al (1966); c Inukai (1962) d Carpenter and Turner (1934); e Grieve and Hey 

(1932-1934); ' NgPh et al (1951); 9 Ray and Rieveschl (1965) 



significant. In order to test this, log I is plotted against the H acidity function so that 
the slope of the straight line is minus unity. The pK BH+ values show a regular variation 
with substituents: electron-withdrawing groups decrease and electron-donating groups 



Table 2. pK BH+ of substituted biphenyl-4-carboxylic acids obtained from the plot 
log / vs H 



Substituent 


r 


c 


m 


c/m 


-P K BH + 


H 


0-999 


6-85 


0-951 


6-90 


6-9 


3'-F 


0-999 


6-68 


0-952 


7-02 


7-0 


3'-Cl 


0-999 


6-72 


0-949 


7-08 


7-1 


3'-Br 


0-995 


6-74 


0-951 


7-09 


7-7 


3'-NO, 


0-999 


6-85 


0-945 


7-25 


7-2 


3'-CH 3 


0-990 


6-50 


0-949 


6-85 


6-9 


3'-OCH 3 


0-994 


6-66 


0-949 


7-00 


7-0 


4'-F 


0-999 


6-59 


0-951 


6-93 


6-9 


4'-Cl 


0-990 


6-66 


0-951 


7-00 


7-0 


4'-Br 


0-999 


6-68 


0-952 


7-02 


7-0 


4'-NO 2 


0-998 


6-97 


0-947 


7-36 


7-4 


4'-CH 3 


0-999 


6-46 


0-950 


6-80 


6-8 


4'-OCH 3 


0-990 


6-15 


0-954 


6-45 


6-4 


2'-F 


0-999 


6-51 


0-949 


6-86 


6-9 


2'-Cl 


0-999 


6-46 


0-946 


6-82 


6-8 


2'-Br 


0-990 


6-43 


0-946 


6-80 


6-8 


2'-NO 2 


0-999 


6-49 


0-952 


7-03 


7-0 


2'-Me 


0-998 


6-36 


0-952 


6-68 


6-7 


2'-OMe 


0-990 


6-20 


0-953 


6-51 


6-5 



pK BH + values are accurate within 0-2 unit; r - correlation coef- 
ficient; m = slope in the plot of log / vs H ; c - intercept in the plot of 
log / vs H 

Table 3. pK. BH + of substituted biphenyl-4-carboxylic acids obtained by Bunnett-Olsi 
treatment. 



Substituent 


r 


[H 2 SO 4 ] at half-protonation 
* %(w/w) -pK BH+ 


H 


0-999 


0-053 


79-33 


6-9 


3'-F 


0-995 


0-052 


80-15 


6-8 


3'-Cl 


0-999 


0-075 


80-56 


6-8 


3'-Br 


0-988 


0-067 


80-64 


6-8 


3'-NO 2 


0-976 


0-079 


81-78 


6-9 


3'-CH 3 


0-998 


0-045 


78-92 


6-6 


3'-OMe 


0-998 


0-052 


80-07 


6-7 


4'-F 


0-969 


0-058 


79-74 


6-7 


4'-Cl 


0-997 


0-074 


80-15 


6-7 


4'-Br 


0-983 


0-065 


80-15 


6-8 


4'-NO 2 


0-982 


0-079 


82-52 


7-1 


4'-CH 3 


0-993 


0-031 


78-59 


6-5 


4'-OMe 


0-978 


0-041 


75-57 


6-2 


2'-F 


0-999 


0-053 


78-84 


6-6 


2'-Cl 


0-997 


0-042 


78-27 


6-5 


2'~Br 


0-987 


0-028 


78-27 


6-5 


T-NO- 


ft-QQ8 


n-rvwi 


n-f7 


fi.7 



Table 4. pKgfj+ of substituted 4-acetylbiphenyl obtained from log/ vs H plot. 



Substituent 


r 


c 


m 


c/m 


~PK-BH + 


H 


0-995 


5-69 


0-909 


6-25 


6-3 


3'-F 


0-990 


6-08 


0-938 


6-48 


6-5 


3'-Cl 


0-999 


6-12 


0-941 


6-50 


6-5 


3'-Br 


0-995 


6-19 


0-953 


6-50 


6-5 


3'-N0 2 


0-999 


6-29 


0-933 


6-74 


6-8 


3'-Me 


0-999 


5-85 


0-967 


6-05 


6-0 


3'-OMe 


0-990 


6-10 


0-960 


6-35 


6-4 


4'-F 


0-999 


5-90 


0-936 


6-30 


6-2 


4'-Cl 


0-998 


5-95 


0-930 


6-39 


6-4 


4'-Br 


0-990 


6-01 


0-940 


6-39 


6-4 


4'-NO 2 


0-999 


6-36 


0-926 


6-87 


6-9 


4'-Me 


0-990 


5-84 


0-974 


6-00 


6-0 


4'-OMe 


0-995 


5-42 


0-984 


5-51 


5-5 


2'-F 


0-999 


5-69 


0-931 


6-11 


6-1 


2'-Cl 


0-999 


5-68 


0-937 


6-08 


6-1 


2'-Br 


0-999 


5-66 


0-933 


6-07 


6-1 


2'-NO 2 


0-990 


6-20 


0-949 


6-53 


6-5 


2'-Me 


0-999 


5-52 


0-936 


5-90 


5-9 


2'-OMe 


0-999 


5-47 


0-974 


5-6 


5-6 



pK BH+ values are accurate within +0-2 unit 



Table 5. pK BH + of substituted 4'-acetylbiphenyls by Bunnett-Olsen treatment. 



Substituent 


r 


[H 2 SO 4 ] at half protonation 
<t> % (w/w) -pK B , r 


H 


0-989 


0-096 


74-0 


5-8 


3'-F 


0-969 


0-067 


75-7 


6-1 


3'-Cl 


0-993 


0-064 


76-1 


6-1 


3'-Br 


0-983 


0-048 


76-1 


6-2 


3'-N0 2 


0-998 


0-073 


78-1 


6-4 


3'-Me 


0-960 


0-036 


72-3 


5-7 


3'-OMe 


0-996 


0-043 


74-8 


6-0 


4'-Fe 


0-953 


0-081 


74-3 


5-8 


4'-Cl 


0-997 


0-076 


75-2 


6-0 


4'-Br 


0-955 


0-064 


75-2 


6-0 


4'-NO 2 


0-970 


0-083 


79-0 


6-5 


4'-Me 


0-992 


0-028 


71-9 


5-6 


4'-OMe 


0-999 


0-017 


68-0 


5-2 


2'-F 


0-997 


0-073 


72-9 


5-7 


2'-Cl 


0-890 


0-065 


72-5 


5-7 


2'-Br 


0-909 


0-060 


72-5 


5-7 


2'-NO 2 


0-965 


0-043 


76-2 


6-3 



Substituent 


By FMMF 
method 


From 
pKa values 


3'-F 


0-107 


0-146 


3'-Cl 


0-122 


0-146 


3'-Br 


0-130 


0-115 


3'-NO 2 


0-255 


0-162 


3'-CH 3 


-0-030 


-0-054 


3'-OCH 3 


0-022 


-0-031 


4'-F 


0-044 


0-085 


4'-Cl 


0-093 


0-154 


4'-Br 


0-102 


0-154 


4'-NO 2 


0-308 


0-323 


4'-CH 3 


-0-065 


-0-039 


4'-OCH 3 


-0-084 


-0-108 



values by dividing the A pKa values by a p value of 1 -32 based on benzoic acid ionization 
in 50 % 2-n-butoxyethanol- water. The new o values thus calculated are given in table 6. 
The Hammett correlations with these values were very poor, the 4'-OMe group 
deviating considerably. Neglecting this group the correlation produces r = 0-871 and 
s = 0-127 in ketones and r = 0-812 and s = 0-1 10 in acids. The poor correlation possibly 
lies in the pKa values of Byron et al (1966) which generated separate lines for 3'- and 
4'-substituted biphenyl-4-carboxylic acids when plotted against ordinary a values. 
When the pK BH+ values are plotted (figure 1) against ordinary Hammett a m and a p 
values respectively for 3'- and 4'-substituents the correlation seems to be much better 
without 4'-OMe, with r = 0-958 and s =-- 0-052 in ketones and r = 0-954 and s = 0-041 
in acids. However, the correlation with a u values calculated (table 6) based on the 
Dewar-Golden-Harris treatment (1971) was satisfactory for all the groups except 4'- 
OMe with r = 0-971 and s = 0-062 for ketones and r = 0-968 and s = 0-045 for the 
acids. This provides an interesting confirmation that the DGH treatment is fairly 
successful for biphenyl (figure 2). 

Being an electron-donating group, 4'-OMe is involved in extended conjugation with 
the protonated carbonyl or carboxyl groups as illustrated in (I). 




This type of direct resonance may be responsible for the departure of the group from 
the Hammett plot. When a + constants of Brown and Okamoto (1958) are used in place 
of ff m and 0p, the correlation is excellent, the 4'-OMe group also falls on the line 
(figure 3). The correlation coefficient is 0-977 with s = 0-053 for ketones and 0-965 with 
<r = 0-067. for acids. The nlot of nK... ?;.<? nKa of the 2'-suhstituted binhenvl-4- 



6-6 



1-6.2 
m 



5.8 



12 



13' 



7.1 



6-9 



6-7 



-0.2 



0.2 



0.6 



Figure 1. pK BH+ us Hammett a plot. 




J FMMF 



0.32 



Figure 2. pK BH + ys T FMMF plot. 



6.6 



s: 

Q. 



5-8 






12 



7-1 



6.9 x 
GO 



6.7 



-0-6 



-0.4 



-0.2 



0-2 



0.4. 



0.6 



Figure 3. pK BH+ as cr + plot 



3. Experimental 

3.1 Preparation of compounds 

All the ketones were prepared as described by Byron et al (1966) and the acids obtained 
by hypochlorite oxidation. The purity of the compounds tested by TLC gave good 
carbon and hydrogen analyses (table 1). 

3.2 Measurement ofpK BH , 

Sulphuric acid (E Merck, AR) was diluted with water and 50-98 % (w/w) solutions were 
prepared. pK BH+ was determined by the procedure adopted by Ananthakrishna Nadar 
and Varghesetharumaraj (1981). A weighed sample of each compound was dissolved in 
85 % (w/w) sulphuric acid- water to give a4xlO~ 4 M stock solution, from which 1 ml 
aliquots was pipetted out into 10 nil volumetric flasks and made up to mark with 
suitable sulphuric acid-water mixtures so as to give solutions of desired acid 
concentrations. The H values were taken from the compilation of Paul and Long 
(1957). Trial experiments indicated that all the ketones and acids were almost 
unprotonated up to 50 % sulphuric acid solution and almost completely protonated in 
95% and above sulphuric acid solution. The uv absorption spectra of the compounds 
were recorded in 50 and 95 % sulphuric acid solutions, to obtain the wavelengths of 
maximal absorption of the unprotonated (A B ) and of the protonated (A BH+ ) forms. The A 
values did not change with change in medium. The extinction coefficients in all the other 
solutions were determined at these two wavelengths. 



7.0 - 



6-9 



6.8 



6.7 



6.6 



6.5 



5.50 



5.58 



5.66 



5.74 



Figure 4. pK BH + vs pKa plot for 2'-substituents. 



The ionization ratio / in each solution was determined as follows: 



where is the molar extinction coefficient in the chosen acid solution, BH+ and B are the 
corresponding values of the completely protonated and the unprotonated forms 
respectively. The ionization ratio is related to H through 

= mH +C. (2) 



From the plot of log / vs H , the H value at half protonation (Ho /2 ) is taken as pK BH+ . 
Following the treatment of Bunnett and Olsen (1965), the pK BH+ values are also 
determined graphically employing equation (3) 



log [BH + ]/[B] + H = 0(H + log CH + ) + pK BH+ (3) 

The intercept in the plot between loerBH + l/rBl +Hn and (H +loeCH + ) gives 



obtained by equation (3) and hence the values obtained from equation (2) are used in 
the correlation. 



Acknowledgement 

NK thanks CSIR, New Delhi for a fellowship. 

References 

Ananthakrishnanadar P and Varghesetharumaraj G 1981 Indian J. Chem. A20 295 

Arnett E M and Anderson J N 1963 J. Am. Chem. Soc. 85 1542 

Brown H C and Okamoto Y 1958 J. Am. Chem. Soc. 80 1979 

Bunett J F and Olsen F P 1965 J. Chem. Soc. Commun. p. 601 

Bunett J F and Olsen F P 1966 Can. J. Chem. 44 1899 1917 

Byron D J, Gray G W and Wilson R C 1966 J. Chem. Soc. (C) 831 837 840 

Carpenter B R and Turner E E 1934 /. Chem. Soc. 869 

Dewar M J S, Golden R and Harris J M 1971 J. Am. Chem. Soc. 93 4187 

Grieve W S M and Hey D H 1932 1933 1934 J. Chem. Soc. 1892 968 1798 

Hammett L P and Deyrup A J 1932 J. Am. Chem. Soc. 54 2721 4239 

Hammett L P and Deyrup A J 1933 J. Am. Chem. Soc. 55 1900 

Inukai T 1962 Bull. Chem. Soc. Jpn. 35 400 

Liler M 1971 Reaction mechanisms in sulphuric acid and other strong acid solutions (New York: Academic 

Press) p. 178 

Long L M and Henze H R 1941 J. Am. Chem. Soc. 63 1939 
NgPh, Buu Hoi, Ng Hoan and Royer R 1951 Bull. Soc. Chem. France 17 489 
Paul M A and Long F A 1957 Chem. Rev. 57 1 

Ray F G and Rieveschl Jr G 1965 Org. Synth. Collog. (London: John Wiley) Vol. Ill p. 23 
Stewart R and Yates K 1958 J. Am. Chem. Soc. 80 6355 
Stewart R and Yates K 1960 J. Am. Chem. Soc. 82 4059 



Effect of substituents on the oxidation of some aikyi-aryl suiphoxides 
by chloramine-T 

K GANAPATHY* and P JAYAGANDHI 

Department of Chemistry, Annamalai University, Annamalainagar 608 002, India 

MS received 4 June 1982; revised 22 April 1983 

Abstract. The oxidation rates of some substituted phenyl methyl suiphoxides with 
chloramine-T have been studied in alkaline and neutral media. OsO 4 is used as catalyst in 
alkaline medium where the meta and para substituents show no effect on the reaction rate. This 
is explained on the basis of isokinetic relationship. In both the media, the orthosubstituents 
show steric effect. 

Keywords. Chloramine-T; isokinetic; kinetics; steric effect; suiphoxides. 



1. Introduction 

The kinetics of oxidation of some substituted phenyl methyl suiphoxides by CAT in 
alkaline medium (Ganapathy and Jayagandhi 1982a) and in buffered ethanol-water 
(Ganapathy and Jayagandhi 1982b) have been studied. In this paper, the substituent 
effects are analysed in terms of steric factor and isokinetic relationship. 

2. Experimental 

Chloramine-T and the suiphoxides were prepared and purified using standard methods. 
Alkaline medium was maintained using NaOH. r-Butanol-water mixture (1 : 1 v/v) was 
used as solvent. For the buffered medium, phosphate buffer (pH 7) was used and 
ethanol-water mixture (1:1 v/v) was used as solvent. Both the kinetics were followed 
iodornetrically. 

3. Results and discussion 

3.1 Alkaline medium 

The observed rate-law for the Os(VIII)-catalysed oxidation of substituted phenyl 
methyl suiphoxides by CAT (Ganapathy et al 1982a) is 

-d[CAT]/d* = ^i. 5 [CAT][SO] 1/2 . 

The mechanism is proposed by assuming the formation of the following cyclic complex 
between CAT and OsO 4 . 
The cyclic structure indicates that the electron density around nitrogen is decreased 




in complexation, weakening the N-C1 bond. Therefore, the interaction with the 
sulphoxides is increased resulting in the formation of chlorosulphonium ion inter- 
mediates. This, on hydrolysis, gives the sulphone. 

The calculated rate constants and the activation parameters for the meta and para 
substituted phenyl methyl sulphoxides are given in tables 1 and 2 respectively. 

A perusal of the rate data indicate that there is no substituent effect on the reaction 
rate. This may be due to two factors (i) a zero-order dependence on the substrate and (ii) 
the experimental temperature being too close to the isokinetic temperature (Leffler 
1955; Leffler and Grumwald 1963; Peterson et al 1961). The first probability is ruled out 
as the present reaction showed a clear dependence (0-5) on the substrate. Hence, the 
only probability is that the experimental temperature should be in the neighbourhood 
of isokinetic temperature. 

A plot of AH* vs AS* gives an excellent straight line with a correlation coefficient (r) 
of 1. The isokinetic temperature obtained from the slope is 305K. It is probable that at 
the experimental temperatures employed (298, 303 and 308K) there will be equili- 
zation of the rates. 

The free energy of activation AG* is equal to (AH* -TAS*) and it follows that if 
there is an exact linear relationship between TAS* and AH*, with unit slope, there will 
be no variation in AG* i.e. all the rate constants are equal. The present reaction series 
showed an excellent relationship between TAS * and AH * with a correlation coefficient 
of 0-998. The slope of the line was 0-984 (nearly unity) which clearly explains that there 



Table 1. Rate constants for Os(VIII)-catalysed oxidation of meta and para substituted 
phenyl methyl sulphoxides by chloromine-T. 
[NaOH] = 0-002 M; [OsO J = 0-0002 M 



fc l . 3 xl0 3 l 1/2 mor 1/2 sec- 1 



Substituent 25 30 35 



None 


4-612 


5-850 


7-086 


p-OCH 3 


4-990 


5-975 


7-372 


p-Cl 


4-557 


5-727 


6-551 


p-Br 


4-756 


5-982 


7-224 


p-CH 3 


4-662 


5-968 


7-503 


p-N0 2 


4-907 


6-085 


7-835 


m-OCHj 


4-618 


5-967 


7-232 


m-Cl 


4-637 


5-778 


7-290 



Substituent 


E 

(kJmor 1 ) 


AH* 

(kJmor 1 
at 30) 


AS* 
(k-T'mor 1 
at 30) 


AG* 
(kJmoP 1 
at 30) 


None 


33-77 


31-25 


-185-2 


87-36 


p-OCH 3 


29-16 


26-64 


-200-2 


87-31 


p-Cl 


28-71 


26-19 


-201-5 


87-25 


p-Br 


32-48 


29-96 


-189-3 


87-31 


p-CH 3 


35-78 


33-26 


- 178-4 


87-31 


p-NO 2 


36-54 


34-02 


-175-7 


87-26 


m-OCH 3 


32-81 


30-20 


-188-2 


87-31 


m-Cl 


33-77 


31-25 


-185-3 


87-40 


m-CH 3 


42-83 


40-31 


- 155-7 


87-48 


m-NO 2 


28-71 


26-19 


-201-5 


87-25 



Table 3. Rate constants for Os(VIII)-catalysed oxidation of ortho-substituted phenyl 
methyl sulphoxides with chioramine-T^ 
[NaOH] = 0-002 M; [OsO 4 ] = 0-0002 M 



Substituent 25 30 35 



None 


4-612 


5-850 


7-086 


Cl 


1-697 


2-041 


2-516 


Br 


1-676 


2-093 


2-771 


OCH 3 


2-146 


2-652 


3-359 


CH 3 


1-915 


2-334 


3-140 


NO 2 


2-441 


3-079 


3-845 


2,6-(CH 3 ) 2 


0-796 


1-134 


1-642 



is exact compensation between TAS* and AH*. 

The rate constants and the thermodynamic parameters for the orthosubstituted 
phenyl methyl sulphoxides are given in tables 3 and 4 respectively. Leffler (1955) 
pointed out that moderate changes in the degree of steric hindrance often do not 
remove a reaction from its isokinetic line but merely move it to a new location on the 
same line. With a considerable increase in steric hindrance in the transition state, a 
combined increase in the enthalpy of activation and decrease in entropy of activation 
are to be expected. Such points lie above the isokinetic line for the other reactions. In 
this study the points corresponding to the orthosubstituted sulphoxides lie above the 
isokinetic line showing steric effect (figure 1). 

2.2 Neutral medium 

The rate-law used to calculate the rate constants k t and k' 2 is 
rorn _ L j_ 



Substituent 


E. 

(kJmol" 1 ) 


(R.J I11U1 

at 30) 


|KJ I11U1 

at 30) 


at 30) 


None 


33-77 


31-25 


-185-2 


87-36 


Cl 


30-83 


28-31 


-203-7 


90-02 


Br 


39-55 


37-03 


- 174-7 


89-95 


OCH 3 


33-49 


30-97 


- 192-7 


89-36 


CH 3 


37-21 


34-69 


-181-5 


89-68 


N0 2 


34-17 


31-65 


- 189-2 


88-98 


2,6-(CH 3 ) 2 


57-41 


54-89 


- 120-8 


91-50 



41 

39 - 
37 

35 

7^ 
E 33 

J 

*l 31 



29 



27 



25 



m-CH, 




- o-CL 



3-OCH 3 
'p-Cl,m-N0 2 



210 200 190 180 170 160 150 
-AS*(J K~ 1 moL~ 1 } 



Figure 1. Plot of AH* vs AS* for the reaction of chloramine-T with substituted phenyl 
methyl sulphoxides in alkaline medium. 



where [SA] is the concentration of p-toluenesulphonamide formed during the reaction. 
/Cj and k 2 are the rate constants for the reaction of .RNHC1 and #NC1 2 with the 
sulphoxides, respectively. The rate constants k 1 and k' 2 are given in tables 5 and 6 
respectively. 

The rate constants of ortho-substituted phenyl methyl sulphoxides clearly reveal that 
the rates are very much lower than those obtained for the corresponding meta and para 
substituted sulphoxides. The steric effect of an ortho-substituent can be estimated by 
calculating the ratio of the rate constants of the ortho and para substituted sulphoxides 
(Capon 1964). The very small relative rate values for methyl and chloro substituents 



k t x 10 4 Imol ' sec * 
Substituent 35 40 45 



None 


46-67 


67-68 


104-20 


OCH 3 


48-32 


53-45 


61-70 




(0-463) 


(0-371) 


(0-322) 


CH 3 


9-26 


10-50 


11-78 




(0-075) 


(0-066) 


(0-068) 


Cl 


1-13 


1-59 


3-43 




(0-069) 


(0-077) 


(0-131) 


Br } 








V 


The reactions are too slow for measurement 


NOj 









The fci(0)/fcj(p) values are given within brackets. 



Table 6. Rate constants for the reaction of RNC1 2 with ortho-substituted phenyl methyl 
sulphoxides. 



/o', xlOMmoP'sec' 1 



Substituent 35 40 45 

None 1678 2069 2265 

OCH 3 1001 1095 1246 

(0-394) (0-349) (0-323) 

CH 3 310-8 4i2-0 486-3 

(0-128) (0-120) (0-122) 

Cl 4-82 6-02 8-95 

(0-012) (0-013) (0-015) 

Br 1 

> The reactions are too slow for measurement 

The k' 2 (o)lk' 2 (p) values are given within brackets. 



show the steric effect. The reactions of CAT with nitro and bromo substituted 
sulphoxides are too slow for measurement. This may be due to the lower stability of the 
transition state because there is greater crowding in the transition state. 

Among the ortho-substituted phenyl methyl sulphoxides, the highest rate was 
observed for o-methoxyphenyl methyl sulphoxide, whose rate constant (k) is higher 
than that for methyl phenyl sulphoxide. Such a high rate deserves special attention in 
view of the neighbouring group effect on the o-QCH 3 group. The presence of an ortho- 
substituent (in this case the reaction centre itself) restricts the free rotation of the 
methoxy group and increases its probability of attaining planarity with the benzene 
ring. Thus there can be enhanced conjugation of the methoxy group with the reaction 
centre. This may explain the higher rate observed for o-methoxyphenyl methyl 



Acknowledgement 

The authors are grateful to the UGC, New Delhi for the award of a fellowship to one of 
them (PJ). 

References 

Capon B 1964 Q. Rev. 18 45 

Ganapathy K and Jayagandhi P 1982a J. Indian Chem. Soc. (communicated) 
Ganapathy K and Jayagandhi P 1982b Int. J. Chem. Kinetics 15 129-139 
Leffler J E 1955 J. Org. Chem. 20 1202 

Leffler J E and Grumwald E 1963 Rates and equilibria of organic reactions p. 324 
Peterson R C, Markgraf J H and Rose S D 1961 J. Am. Chem. Soc. 83 3819 
Ruff F, Kapovits I, Rabai J and Kucsman A 1978 Tetrahedron 34 2767 

Satyanarayana P V V 1977 Kinetics of oxidation and reduction of some aryl methyl sulphoxides Ph.D. Thesis, 
Annamalai University 



Spectroscopic studies of the electron donor-acceptor interactions of 
aromatic hydrocarbons with tetrachlorophthalic anhydride 

P C DWIVEDP and AVANIJA GUPTA 

Department of Chemistry, Harcourt Butler Technological Institute, Kanpur 208 002, India 
MS received 9 February 1983; revised 29 April 1983 

Abstract. Electron donor-acceptor (EDA) interactions of tetrachlorophthaiic anhydride 
(TCPA) with benzene, p-xylene, mesitylene, fluorene, f-stilbene and pyrene have been 
investigated by spcctroscopic technique. The spectroscopic and thermodynamic parameters of 
the complexes formed are reported. The enthalpies of formation range between 0-2 to 
5-0 kcal mole" 1 . Among all the donors studied, pyrene appears to be the strongest electron 
donor towards tetrachlorophthalic anhydride. 

Keywords. Tetrachlorophthalic anhydride; electron donor acceptor complexes; formation 
constants; spectroscopic studies. 



1. Introduction 

Our interest in tetrachlorophthalic anhydride (TCPA) as electron acceptor arose from 
the observation that the electron affinity of TCPA (1-8 eV) is comparable to that of 1,3,5- 
trinitrobenzene (1-86 eV), which is known to be a good electron acceptor. Inspite of this, 
relatively little attention has been focussed towards investigating the electron donor- 
acceptor (EDA) complexes of TCPA (Dwivedi and Banga 1980a). Stable molecular 
complexes of TCPA with aromatic hydrocarbons and aza-aromatics are known 
(Czekalla 1956) to form. Formation constants of 1 : 1 EDA complexes of polynuclear 
aromatic hydrocarbons with TCPA have been determined by Chowdhury and Basu 
(1960). Dwivedi and Banga (1980a) have reported the enthalpies of formation of a 
number of EDA complexes formed between aromatic hydrocarbons and TCPA. The 
/iv CT -ionization potential plot was found to be linear. As part of an extended 
programme to study the EDA complexes of TCPA, we have presently investigated the 
interaction of TCPA with benzene, p-xylene, mesitylene, fluorene, t-stilbene and pyrene. 

2. Experimental 

Benzene, p-xylene (BDH, AR) and mesitylene (SRL) were purified by fractional distillation. 
Fluorene, t-stilbene and pyrene (supplied by Aldrich Chemical Co., USA) were purified 
by recrystallization from alcohol. TCPA was repeatedly crystallized from benzene until 
its absorption spectrum in CC1 4 showed no further change on successive crystallization. 
CC1 4 (E. Merck, G. R.) was dried and distilled before use. 



30 



26 



? 22 



18 



14 



12 




8 



12 



18 



(S-l-H-C) x 10 



Figure 1. Modified Scott equation plot for the fluorene-TCPA complex at a. 21C and b. 
31C. (S and H are the initial concentrations of the donor and acceptor, C is the concentration 
and A is the absorbance of the complex and I is the optical path length). 

fitted with variable temperature cell compartment using matched silica cells of 1 cm 
path length. Equilibrium constant of formation, K, and molar extinction coefficient, , 
of the EDA complexes were determined employing the modified Scott equation (Scott 
1956) by measuring the absorbance (at the A max of the charge-transfer band) of a series 
of solutions with varying donor concentration and a fixed TCPA concentration. In 
evaluating K, Person's criteria regarding donor concentration were satisfied (Person 
1965). The K and e values of the complexes have an uncertainty of less than + 10 % as 
can be seen from figure 1 for the Fluorene-TCPA system. The enthalpies of formation, 
AH, of the EDA complexes were evaluated from the equilibrium constants at different 
temperatures in the range 10-40C. Freshly prepared stock solutions of the donor and 
acceptor were used in all the measurements. 

3. Results and discussion 

The charge-transfer bands of all the EDA complexes appear in the region of the donor 

absorption and therefore the charge-transfer band maxima (A CT values) were obtained 

by difference spectra (figure 2). The results of the present investigation are summarised 

in table 1. The A CT values for the EDA complexes of TCPA with z-stilbene and pyrene agree 

well with those reported by Chowdhury and Basu (1960). However, the K and e values 

in these systems differ from our data. We believe that the present values are more 

reliable since the donor concentration employed was sufficiently high and satisfied 

Person's (1965) criteria. Mulliken (1952) pointed out that the lower the ionization 

potential, higher will be the stability (as measured by K and AH) of the EDA complex. 

This is found to be true from our data in table 1. While uncertainties in the 

determination of A// in systems with small K values would be large, the AH values are 

estimated to be well within 10% even after accounting for all possible sources of 

error. These AH and K values for the complexes studied are comparable to other 

Tr-donor-Tt-acceotor svstems (Rao et al 1971). indicating that TCPA acts as a rc-acceotor in 



0.18 - 



fe 0.12 - 



0.06 - 




340 



360 
Wavelength(nm) 



380 



Figure 2. Electronic absorption spectra of a. benzene (0-4 M) + TCPA (4-0 x 10~ 4 M) and 
b. p-xylene (0-4 M) + TCPA (4-0 x 10" 4 M) in CC1 4 solution. 



Table 1. Spectroscopic and thermodynamic data for EDA complexes of aromatic hydro- 
carbons with TCPA in CCL solution. 



3n donor 


1, 

(eV) 


A CT 
(nm) 


K* 

(ImoP 1 ) 


e 
(lmor l cm~ l ) 


-AH 

(kcalmol' 1 ) 


Av 1/2 
(cm' 1 ) / 


ne 


9-24 


350 


1-5 0-1 


666 40 


0-24 0-02 


1243 0-003 


ne 


8-44 


350 


2-2 + 0-2 


1363 + 40 


0-46 0-03 


1787 0-011 


/lene 


8-39 


350 


3-5 0-2 


150050 


1-60-1 


2152 0-014 


:ne 





350 


6-7+0-2 


1177 50 


2-6 + 0-2 


812 0-004 


:ne 


7-99 


350 


7-7 0-1 


1077 40 


3-6 + 0-2 


1124 0-005 





7-82 


425 


10-4 0-1 


1300 50 


5-0 0-2 


3323 0-043 



21; data are given at one temperature only for the sake of brevity. 



i (Av 1/2 ), e and oscillator strength (/') data can be analysed if one carefully chooses 
on donors which are similar in structure. Thus, if the donors in table 1 are 
dered in three separate groups of methyl substituted (group 1), phenyl substituted 
ip 2) and polynuclear hydrocarbons (group 3), the Av 1/2 order is: 

mesitylene > p-xylene > benzene 
t-stilbene < benzene 
pyrene > fluorene < benzene 

e Av 1/2 values increase with increasing complex strength (as measured by - AH) 
the exception of benzene. This direct relationship between Av 1/2 and the strength 
s complexes was observed earlier (Dwivedi and Banga 1980b; Dwivedi et al 1982) 
#as attributed to the large resonance interaction in the complexes, 
e and /values of the EDA complexes provide a measure of the intensity of the 
^e-transfer (CT) band. Based on these data in table 1, the CT band intensity order is: 

mesitylene > p-xylene > benzene 



It is, thus, seen that the intensity of the cr band increases as the strength of interactio 
increases in electron donors of all the three groups. This observation is similar to th* 
reported by Daisey and Sonnessa (1972) and Dwivedi and Banga (1980b,c) for 
number of EDA complexes. 

Acknowledgement 

The award of a Junior Research Fellowship (to AG) by the CSIR, New Delhi, is grateful! 
acknowledged. 

References 

Chowdhury M and Basu S 1960 Trans. Faraday Soc. 56 335 

Czekalla J 1956 Naturwissenschaften 43 476 

Daisey J M and Sonnessa A J 1972 J. Phys. Chem. 76 1895 

Dwivedi P C and Banga A K 1980a Indian J. Chem. A19 158 (and reference cited therein) 

Dwivedi P C and Rao CNR 1970 Spectrochim. Ada A26 1535 

Dwivedi P C, Gupta A and Banga A K 1982 Curr. Sci. 51 651 (and references cited therein) 

Dwivedi P C and Banga A K 1980b Indian J. Chem. A19 908 

Dwivedi P C and Banga A K 1980c J. Inorg. Nucl. Chem. 42 1767 

Mulliken R S 1952 J. Am. Chem. Soc. 74 811 

Person W B 1965 J. Am. Chem. Soc. 87 167 

Rao CNR, Bhat S N and Dwivedi P C 1971 Appl. Spectrosc. Rev. 5 1 

Scott R L 1956 Reel. Trav. Chim. Pays-Bas Belg. 75 787 



A zero differentia! overlap study of chemical binding in ammonia- 
borane and carbonyl-borane 

DULAL C GHOSH 

Department of Chemistry, University of Kalyani, Kalyani 741 235, India 
MS received 8 May 1982; revised 12 April 1983 

Abstract. A CNDO/2D study of the charge distribution obtained through Mulliken population 
analysis in the ground state of the title compounds shows that the features of charge 
distribution found by several ab initio calculations are fairly well reproduced by this method. 
The one-particle density, the interference density at the mid-point of the bond axis and the 
kinetic part of the interference energy calculated through the deorthogonalized density 
matrices over a wide range of intermolecular separation between the donor and the acceptor 
show that the one-particle density and the interference density steadily grow with decreasing 
internuclear separation, while the kinetic interference energy starts with negative value at large 
distance, then decreases and passes through a minima near but above the equilibrium distance 
and then increases rapidly below it conforming to the characteristic general behaviour of the 
kinetic component of Morse curve. The orbital pairwise interference density and the 
corresponding kinetic energy components reveal that the orbitals involved in the covalent 
binding are a 2p AO of B and 2S and cr 2p AO of N and C atoms in H 3 B-NH 3 and H 3 B-CO 
respectively. 

Keywords, zoo; deorthogonalization; one-particle density; interference density; kinetic 
interference energy; ammonia-borane; carbonyl-borane. 



1. Introduction 

According to the conventional concept of chemists, ammonia borane (or borazane), 
H 3 B-NH 3 , and carbonyl borane, H 3 B-CO, are formed by the sharing of a lone pair of 
electrons belonging to the donor groups (NH 3 , CO) with the vacant orbital of the 
acceptor group (BH 3 ). The newly formed B-L bond (L is any ligand) is of pivotal 
importance to account for the stability and characteristics of such boron coordination 
compounds. A number of theoretical calculations probing the various aspects of these 
two molecules have appeared (Das 1957; Hoffmann 1964; Veillard et al 1967; Moireau 
and Veillard 1968; Peyerim HofF and Buenker 1968; Armstrong and Perkins 1969; 
Lloyd and Lynaugh 1970, 1972; Frost 1970; Shillady et al 1971; Gordon and England 
1972; Palke 1972; Bach et al 1973; Fujimoto et al 1974; Kato et al 1974; Purcell and 
Martin 1974; Runtz and Bader 1975; Dill et al 1975; Umeyama and Morokuma 1976; 
Ermler et al 1976; Ha 1976; Datta et al 1977; Redmon et al 1979; Zirz and Alrichs 1981). 
The probing into the nature of the newly formed bond of these compounds has so far 
been in terms of Mulliken (1955) population analysis which can make only a qualitative 
prediction as to the distribution of electrons and the binding effect. Ruedenberg (1962) 
suggested an energy analysis, based on density matrix properties, which can extract 
significant energetic information that provides a quantitative basis of reasoning about 
the contributory factors to chemical binding. Ruedenberg concluded that the 



V/liV/ gjj* 5 V V/t V/ 5.7 l 1 Iw w v AJ.Vf VW IVw^Jl W tr XT 

a number of molecular system (Layton Jr and Ruedenberg 1964; Rue and Ruedenberg 
1964; Edmiston and Ruedenberg 1964; Popkie and Moffatt 1968; Fienberg et al 1970). 
Wilson and Goddard (1970, 1972) critically analyzed the role of kinetic energy in 
covalent binding and concluded, from the analysis of GI functions of molecules, that 
lowering of a non-classical 'exchange kinetic energy' is responsible for chemical binding 
and this 'exchange kinetic energy' is related to the interference kinetic energy of 
Ruedenberg. Recently, a number of reports have appeared on the calculation of 
interference density and the kinetic interference energy (Datta 1976; Datta and Datta 
1978), but some fundamental discrepancies are evident in this work (vide infra). 

The purpose of the present investigation is to extract information as to whether 
Ruedenberg's type of binding analysis can be attempted by semi-empirical zoo 
methods like CNDO/2 (Pople et al 1965). The CNDO/2 method, inspite of its success in 
correlating a number of molecular properties with experimental results (Pople and 
Gordon 1967) and its easy applicability to large molecules, especially, the boron 
coordination compounds (Labarre 1978), calculates unrealistic charge distribution 
(Shillady et al 1971) because, the basis functions of this formalism are virtually a set of 
orthogonalized AO's (Pople and Segal 1965). 

The spinles one-particle (electron) density p(R) at a position R for a closed shell 
molecular system expanded in terms of AO's takes the form 

*X (1) 

where P(r, s) are the elements of bond order matrix and 0,-'s are the basis set. Integration 
of p(R), over all space, for methods that include overlap, gives the population 
breakdown 



atoms atoms atoms 

= Z PS+ Z Z PAB. (3) 

A A B> A 

where AT is the total number of electrons and S(r, s) is the overlap integral between AO's 
r and s, and P N A and P AB are the net atomic and overlap populations respectively, 

(r,4 (4) 

s)S(r,s), (5) 



but for zoo methods, (2) takes the form 

atoms A 



x= Z Iffcr). (6) 

A r 

Ruedenberg suggested a partitioning of the electron density p(R) into the quasi- 

n^J in t/arfronr>> rlanoi+ir nl 



N= p(R)dR = p d (R)dR, (8) 

v 



Q. (9) 

sing the conditions (8) and (9), and applying the series of approximation regarding 
iteracting orbital pair (Ruedenberg 1962; Edmiston and Ruedenberg 1964) the 
ila of interference density has been derived as 

p'= p '(AaBb)= P(AaBb)(AaBby, (10) 

Aa, Bb Aa, Bb 

; Aa, Bb are AO's centred on atoms A and B respectively, P(Aa Bb} is the element 
nd order matrix and (AaBby is the orbital interference density for the orbital 
Aa, Bb) 

(AaBby = Aa-Bb~%S(AaBb)(Aa 2 + Bb 2 ), (11) 

: S(Aa Bb) is the overlap integral of the AO pair. To measure the degree to which 
bital (Aa) is effective in creating interference effect, Ruedenberg, similar to that of 
eeny (1951) and Mulliken (1955), partitioned the population q(Aa) of the orbital 
valence inactive', P(Aa), and 'valence active', v(Aa\ parts. 



v(Aa), (12) 

P(Aa) = P(AaAa), (13) 

a Bb) S(Aa Bb). ( 1 4) 



Bb 

elative size and sign of v(Aa) determines the binding capacity of the orbital (Aa). 
w under a zoo formalism, the overlap distribution (Aa-Bb) and the overlap 
al S(AaBb) are all zero thereby excluding any possibility of inclusion of 
erence effect in CNDO/2 method, and the electron density is given by the quasi- 
cal density only and the valence activity (equation (14)) is zero. For the same 
n, overlap population, although used as bond index (Fischer and Kollmar 1970; 
ison and Seltzer 1971; Corre 1981), cannot be extracted from a CNDO/2 calculation 
i the total charge density is divided out formally among the atoms alone 
tion 6), and under such a formalism, the identification of bond is not straight 
ird (Mclver et al 1971; Driessler and Kutzelnigg 1977). 

us it is apparent that the CNDO/2 method is not at all suitable for the analysis of 
:ular binding whatsoever, and the fundamental discrepancies apparent in Datta 
), and Datta and Datta (1978) may now be mentioned (i) these authors have 
lated interference density and related properties through the localized molecular 
Us. But it is not a priori clear how the interference effect which arises due to the 
like behaviour of fundamental particles (Ruedenberg 1962) is incorporated in a 



Ruedenberg (1964) studied the chemical binding in the same system by the same 
method using the same wave function of Ransil. But it is apparent that the 
corresponding values of Datta are different from that of Layton Jr and Ruedenberg, and 
the same wavefunction (evaluated at the equilibrium distance) has been used over a 
range of 21 distances of internuclear separation the reason for which are not mentioned 
in the paper (iii) Datta and Datta (1978) have calculated interference density at the 
bonding region through CNDO/I functions. But interference effect cannot arise here 
because of the orthogonal nature of the basis functions, which cannot be changed by 
the unitary transformation converting the delocalized set into a localized one. Thus, 
the work is not consistent with the basic philosophy of CNDO/2 formalism. 

However, Shillady et al (1971) noted that a CNDO/2D formalism, in which the orbitals 
from the CNDO/2 calculations are deorthogonalized by Lowdin (1950) transformation, 
gives charge distribution that follows closely the trend of ab initio calculations, and 
Mclver et al (1971) noted that the charge density distribution found by such a method is 
comparable with experimental results. Since the basis set is not orthogonal, the overlap 
distribution (Aa Bb) and the overlap integral S (Aa Bb) are no longer zero and the 
inclusion of interference effect is theoretically justified in CNDO/2D formalism. We have 
therefore, undertaken a CNDO/2D study of charge distribution, in terms of orbital and 
overlap population, in H 3 B-NH 3 and H 3 B-CO as compared to the available ab initio 
results. We have also computed one-particle density, interference density at the bond 
mid-point and the kinetic part of the interference energy, the bond population as 
functions of internuclear separation between the donor and the acceptor to study the 
variation of covalent binding with internuclear distance by the same method. The two- 
centre energy term, E A _ B , and one of its components for the new bond are computed, to 
use as bond index, by CNDO/2 energy partitioning. 



2. Method of computation 

The geometric parameters of the donors, the acceptor and the adducts are optimized by 
CNDO/2 method using standard parameters and STO basis set. The overlap and coulomb 
integrals are evaluated from the explicit expressions derived by Roothaan (1951). The 
B-L bond is made to coincide with Z (C 3 ) axis of the coordinate system. The CNoo/2 
density matrix is deorthogonalized according to Lowdin's prescription (Lowdin 1950) 

P' = S~ 1/2 PS- 1/2 (15) 

where P and P' are the density matrices corresponding to orthogonal and non- 
orthogonal bases, and S is the overlap matrix over Slater orbitals respectively. P' is used 
to calculate charge distribution through equations (4) and (5). The computed orbital 
populations, and CNoo/2 vis-a-vis CNDO/2D charge distribution of all the chemical 
species at CNDO/2 equilibrium geometry are shown in table 1 and figure 1 respectively. 
The orbital pairwise decomposed B-N, B-C and C-O bond populations are shown in 
tables 2-4 respectively. The acceptor and the donor moieties are kept at geometry to 
which they are reorganized in the adducts and a series of wavefunctions are calculated 



/ I 



' a v 
H I H 

H + 0-113A 



>0-20d6 
B 



-00682 
(C 3V ) 



0-2408 
B. 

\08222 

\ 
\ 

H H 

-00803 
(C 3V ) 



0-0828 
r 



-00828 



. 0-9791 
1-02050 -0-2050 



-0-1139 
H 



+ 0-1556 



+0-010 



(staggered ) 



+ 00190 
H 

H^V-Q.3833 +0-4887 -0-1624 

B C =O 



0-2090 
H 



(staggered) 



hO-0095 

H 
\ 
\ 



0-7836 



+0-3487 -0-2620 
C ..." O 



1-0024 



Figure 1. Charge distribution in the donors, the acceptor and the adduct. 



ond mid-point, and T 1 , the kinetic part of the interference energy, is calculated by the 
>rmula 



r 
T' = \fp'(R)dR 

J 



AaBb 



(AaBb)dR 



(16) 



AaBb 



here f is the operator for kinetic energy and the integral T(AaBby is the kinetic 
icrgy of the orbital interference density (equation (11)). 

= [T(AaBb)-$S(AaBb){T(AaAa)+T(BbBb)}~\, (17) 



AO 


Donors and 
Acceptors 


Adducts 




BH 3 (C 3 J 


BH 3 -NH 3 


BH 3 -CO 


B 2s 


1-0488 


0-9361 


0-9144 


B 2p ,, B 2py 


0-8272 


0-7771 


0-8143 


B ap, 


0-0560 


0-4222 


0-5680 


HJ^B) 


1-0803 


1-1386 


0-9905 




NH 3 






N 2i 


1-6311 


1-5057 




N 2pi , N 2py 


1-0244 


1-1170 




N 2pi 


1-6600 


1-5588 




H U (N) 


0-8866 


0-7910 






CO 






C 2s 


1-7416 




1-3980 


C 2p;( , C 2py 


0-5453 




0-6435 


C 2p 


0-9627 




0-9664 


oj 


1-8633 




1-8585 


O 2pi , O 2py 


1-4547 




1-5350 


ap] 


1-4322 




1-3336 



Table 2. Breakdown of the B-N bond population (BH 3 -NH 3 ). 



Orbital pairs 



Populations 



B 2s /N 2s 

B 2s /N 2pt 

B 2PI /N 2S 



-0-0125 
0-0658 
0-1925 
0-2050 
0-0150 
0-0150 



Table 3. Breakdown of the B-C bond population (BH 3 -CO). 



Orbital pairs 



Populations 



B 2s /C 2s 

B 2l /C 2pr 

B 2PI /C 2S 



-0-0077 
0-0556 
0-3426 

0-1677 
0-1127 
0-1127 



where the matrix elements are given by 

r 
T(AaBb) = \Aa(R)fBb(R)dR, 



(1 



Orbital pairs 



Populations 





(a) 


(b) 


C 2s /0 2i 


-0-2110 


-0-0992 


C 2s /O 2ps 


0-0753 


0-2267 


C 2 p r /O 2 's 


0-0706 


-0-0464 


C 2p '/O-, p 


0-4221 


0-3804 


n(C 2 Jo 2px ) 


0-3110 


0-2704 


*(C 2py /0 2py ) 


0-3110 


0-2704 



(a) Free CO at the same geometry as in H 3 B-CO;(b) CO 
moiety in H 3 B-CO. 



T(BbBb) = \Bb(R)fBb(R)dR. 

J 



(20) 



a Bb) elements are computed through the explicit formulae of Roothaan (1951). 
In computing p the entire basis set is included, but in computing p 1 the principal 
interacting orbitals of B, N and C only are .considered. The two-centre energy terms 
E A _ B for bonded interaction are a measure of the strength of chemical bond and have 
five components (Fischer and Kollmar 1970) 



= R 



-4-F N 

^ ^ 



(21) 



of which E AB , the contribution of resonance integrals to the energy of the A-B bond, is 
the main feature of the covalent bond and closely correlates with the E AB values. The 
E AB and its E AB components are computed for each B-N and B-C separation. The 
computed values of/?, p',T', E AB , E AB and P AB are shown in tables 5 and 6 for H 3 B-NH 3 
and H 3 B-CO respectively. The orbital pairwise decomposition of p 1 and the 
corresponding T' are shown in tables 7 and 8 for B-N and B-C bonds respectively. 



Table 5. p(R), p' (R), T 1 , E A _ B , 
B-N separation in borazane. 



'he bond populations at several distances of 



r(B-N), (^ 


^) P(R] 


p'(R) 


r' M.) 


E A , B (a.u.) 


EA-B (a.u.) 


B-N bond 
population 


1-165 


0-4237 


0-0654 


-0-0049 


-0-7657 


-1-4412 


0-4222 


1-365 


0-2837 


0-0537 


-0-1690 


-0-8719 


-1-1619 


0-4976 


1-565 


0-1885 


0-0419 


-0-2603 


-0-8119 


-0-9133 


0-4806 


1-765 


0-1228 


0-0311 


-0-2913 


-0-6765 


-0-6914 


0-4189 


1-965 


0-0771 


0-0216 


-0-2744 


-0-5172 


-0-4984 


0-3355 


2.165 


0-0461 


0-0140 


-0-2281 


-0-3196 


-0-2921 


0-2482 


3-50 


0-0426 


0-0001 


-0-0068 


-0-0136 


-0-0060 


0-0051 



r(B-C) (A) p(R) 



p l (R) 



/1 _ B (a.u.) 



B-C bond 
EJ4_ B (a.u.) populations 





1-04 
1-24 
1-44 
1-64 
1-84 
2-04 
3-50 


0-3558 0-0810 0-1031 -1.2171 
0-2692 0-0579 -0-0416. -1-3772 
0-1998 0-0432 -0-1434 -1-2932 
0-1456 0-0320 -0-1962 -1-0890 
0-1033 0-0237 -0-2122 -0-8492 
0-0714 0-0170 -0-2002 -0-6231 
0-0023 0-0003 -0-0101 -0-0134 


- 2.0305 
- 1-6987 
- 1-3804 
- 1-0785 
-0-8080 
-0-5807 
-0-0137 


0-8202 
0-8705 
0-7836 
0-6418 
0-5030 
0-3797 
0-0102 


R = mid point of r(B-C). 

Table 7. The orbital pair-wise breakdowns of the interference density and the correspond- 
ing interference kinetic energies at several distances of B-N separation in borazane. 


r (B-N), 
(A) 


PUW 


P2s/2p(R) P2p/2s(R) P2p!2p(R) T 2s /2s * 2s/2p 


T 1 

1 2pj2s 


T i 

1 2p/2p 


1-165 
1-365 
1-565 
1-765 
1-965 
2-165 
3.50 


-0-0055 
-0-0022 
-0-0005 
0-0002 
0-0004 
0-0003 



0-0036 0-0110 0-0563 0-0417 -0-0208 
0-0061 0-0096 0-0403 0-0178 -0-0442 
0-0057 0-0078 0-0289 0-0046 -0-0509 
0-0045 0-0061 0-0203 -0-0019 -0-0498 
0-0032 0-0044 0-0137 -0-0045 -0-0429 
0-0021 0-0030 0-0086 -0-0051 -0-0338 
0-00002 0-00003 0-00007 -0-0002 -0-0009 


-0-0576 
-0-0779 
-0-0862 
-0-0841 
- 0-0745 
-0-0598 
-0-0017 


0-0318 
- 0-0646 
-0-1278 
-0-1556 
-0-1525 
-0-1294 
-0-0040 



(R) = mid point of r(B-N). 



Table 8. The orbital pair-wise breakdowns of the interference density and the correspond- 
ing interference kinetic energies at several distance of B-C separation in carbonyl borane. 



r(B-C), 

(A) 


,Uw 


/.Utff) 


PW/O 


,Uw 


rU 


T 1 

1 2s/2p 


T 1 

1 2p/2s 


T 1 

1 2p/2p 


1-04 


-0-0052 


0-0039 


0-0198 


0-0625 


0-0388 


-0-0171 


-0-0552 


0-1366 


1-24 


-0-0018 


0-0034 


0-0155 


0-0408 


0-0133 


-0-0201 


-0-0815 


0-0468 


1-44 


-0-0002 


0-0032 


0-0127 


0-0274 


0-0017 


-0-0245 


-0-0990 


-0-0216 


1-64 


0-0004 


0-0026 


0-0104 


0-0186 


-0-0032 


-0-0243 


-0-1055 


-0-0633 


1-84 


0-0005 


0-0021 


0-0082 


0-0128 


-0-0055 


-0-0231 


-0-1018 


-0-0817 


2-04 


0-0006 


0-0017 


0-0062 


0-0087 


- 0-0065 


-0-0203 


- 0-0904 


-0-0831 


3-50 


0-00002 


0-00003 


0-0001 


0-0001 


-0-0005 


-0-0009 


-0-0040 


-0-0047 



(R) = mid point of r(B-C). 



3. Results and discussion 

The HBH angle reorganizes to 110 in both the complexes. The B-H bond length is 



umauicu uy acvciai nun-cuipiiii/cu iiicuiuuis ^nLiuiauuiig aim rciKins li/O;/, jprmier ei al 

1976; Umeyama and Morokoma 1976). The HlSTH angle changes from 105 to 109-25, 
and the N-H bond length (1-07 A) remains unchanged, on adduct formation. The 
reorganization energy of NH 3 is negligible (0-0019 a.u). The optimized B-N bond 
length 1-565 A, is in very good agreement with its experimental value 1-56 A (Shore and 
Parry 1955; Hughes 1956; Lippert and Lipscomb 1956). The optimized B-C bond 
length is 1-44 A which is also in good agreement with its experimental value (Herzberg 
1966). The C-O bond stretches from 1-19 to 1-211 A and the reorganization energy of 
CO is negligible (0-0019 a.u). The staggered form of borazane is the stabler one. 

3.1 Mulliken population analysis 

3.1a Ammonia borane: From figure 1 we see that the CNDO/2Z) formal charges on B in 
BH 3 (C 3w ) is positive and that on N in NH 3 is negative and these formal charges do not 
change appreciably on molecular formation although, 0-3284 a.u of charge has been 
formally transferred from the donor to the acceptor. But the formal charges on all the H 
atoms changes significantly on adduct formation. The charge density on H attached to 
B changes from - 0-0803 a.u to 0-1386 a.u and that on H attached to N changes from 
4- 0-1 134 a.u to + 0-2090 a.u. Since B is nearly neutral and N is strongly negative in the 
complex, the charge rearrangement contradicts the classical belief of the formation of 
the dative bond by the donation of lone pair of electron of N to the empty orbital of B. 
From table 1 we see that the population in B 2pz orbital increases considerably and those 
in all the other B orbital decreases, while in N, the charge donating atom, the 
populations in 2s and 2p z orbitals decrease and those in 2p x and 2p y orbitals, increases. 
The overlap population of the B-N bond is surprisingly low compared to those of B-H 
and N-H bonds. Table 2 demonstrates that the B 2s /N 2s overlap population is a 
negative number (antibonding) and the B-N bond is essentially constructed from 2p a 
orbital of B and 2s and 2p v orbitals of N, and the major contributors are the 2p a orbital 
pair. Thus the charge transfer originates mostly from the N 2pz orbital to the B 2pz orbital 
and a charge readjustment finally takes place between the two central atoms and the H 
atoms bound to them to make N atom strongly negative and B atom virtually 
uncharged. This same pattern of charge distribution and bond formation in this 
molecule were found by several ab initio calculations (Veillard et al 1967; Moireau and 
Veillard 1968; Armstrong and Perkins 1969). On the other hand, CNDO/2 method over- 
estimates the net transfer of charge (0-4765 a.u) and gives positive charge on N and 
negative charge on B (figure 1), which was also noted by Shillady et al (1971). Moreover, 
the bond formation cannot be studied by CNDO/2 method. 

3. 1 b Carbonyl borane: Figure 1 demonstrates that the formal charges on C of CO and 
B of BH 3 are positive. Thus the formation of BH 3 -CO involves the joining of two 
positively charged fragments. The net balance of charge transfer is 0-08 a.u. of electronic 
charge from CO moiety to the BH 3 moiety. These two facts suggest that, during the 
process of bond formation, both the interacting groups act in a concerted way to donate 
and accept charge. On adduct formation, the B atoms are formally negatively charged 
while the H atoms are positively charged, and the formal positive charge on C and 
negative charge on O are increased. Table 1 demonstrates that the population in B(2p z ) 



2s and 2p a orbital of C. Thus the charge donation stems from the 2s and 2p r orbitals of C 
into the 2p r orbital of B and as a final adjustment charge migrates from O(2s) and 
O(2p z ) orbitals into C(2s) and C(2p z ) orbitals. The amount of charge migrated from 
B(2s) orbital must be donated to the H atoms. Since the magnitude of Tc(B-C) 
population is considerable and the electron density around H atoms is decreased, the 
back-donation of charge from BH 3 to CO must be taking place through the 
'hyperconjugative' mechanism as suggested by Graham and Stone (1956). The pi- 
orbitals of B, C and O atoms have simultaneous overlap with the group orbitals of three 
H atoms to drift charge around H towards C and O. Thus Mulliken's categorisation of 
CO as an 'amphodonor' is also supported by this CNoo/2-JD calculation. 

One more interesting point to note is the dramatic increase in C-O bond population 
on adduct formation (figure 1), which may be offered as a natural explanation of the 
spectroscopic observation of the increase in CO stretching frequency on complex 
formation with BH 3 (Bethke and Wilson 1957; Taylor 1957; Sundaram and Cleveland 
1960; Cowan 1949, 1950). The breakdowns of the bond populations of the free CO at the 
same geometry as in the complex, and of CO in the complex into orbital pairwise 
components (table 4) show that the observed increase in the C-O bond population 
arises from the reduction of the large antibonding C 2s /O 2s population and from an 
increase of C 2s /O 2pz population on complex formation. This pattern of charge 
redistribution within the donor and the acceptor and the mechanism of bond formation 
were also observed by several ab initio calculations (Armstrong and Perkins 1969; 
Runtz and Bader 1975; Kato et al 1974; Ermler et al 1976; Umeyama and Morokuma 
1976). Although the gross patterns of charge rearrangement are the same in both 
CNDO/2 and CNDO/2-D methods, the former over-estimates the net transfer of charge 
(figure 1). 

3.2 The one-particle density, the interference density and the kinetic interference energy, 
and other bond indices. 

The quantities in tables 5 and 6 have a common trend of variation. The binding 
interaction starts at a sufficiently large R, the distance of internuclear separation 
between the atoms forming the new bond. The bond-populations, one-particle density 
and the interference density at the bond mid-point steadily grow as R decreases, 
indicating a piling of charge density in the bonding region in both the systems. The E^ B , 
the resonance integral, steadily decreases with decrease of R, while E AB , the energy of 
the newly formed bond, closely correlates with the A-B bond-population and passes 
through a minima at shorter R where the latter is maximum. It is further noted that the 
sum of the reorganization energies of the donor and the acceptor is over-compensated 
by the energies of the newly formed bond in both the systems over a considerably large 
range of internuclear separation. 

The kinetic interference energy T 1 starts with a negative value at large R and sharply 
decreases with the decrease in R, and then rapidly increases below R e , the equilibrium 
R, for both the systems under study. However, IT 7 remains negative over the entire range 
of study in H 3 B-NH 3 , and becomes positive at shorter R values below R e in H 3 B-CO. 



0-10 

0-05 

0-0 

-0-05 

-0-10 

-0-15 

-0-20 

-0-25 

-0-30 



o Carbonyl borane 
Ammonia borane 



1-00 1-20 UO 1-60 1-80 2-00 2-20 2-40 2-60 2-80 3-00 3-20 3-403-50 



Figure 2. Plot of the kinetic interference energy as a function of internuclear distance. 



general pattern of the kinetic component of Morse curve (Slater 1933; Eyring et al 
1944). According to the kinetic component of the Morse curve, kinetic energy decreases 
as R decrease and then increases rapidly as the nuclei are brought closer together. 
Ruedenberg (1962) argued that such lowering of kinetic energy comes through 
interatomic interference, and he conjectured that due to cluster promotion atoms form 
the promotion states in which potential energy is decreased and kinetic energy is 
excessively increased leading to an overall promotion, and subsequent interference 
causes a large compensating lowering in kinetic energy. At larger R, cluster promotion 
is less important and pure interference effect is possible to lower the kinetic energy, but 
at shorter R, particularly at and around R e , cluster promotion is important and 
therefore kinetic energy should start increasing. Although there is no scope of inclusion 
of cluster promotion in the present calculation, the general trend of variation of T' 
values with JR is consistent with the experimental curve of variation in kinetic energy as 
function of internuclear separation (the kinetic component of Morse curve). 

The interference density of 2s(B)/2s(N) orbital pair is positive (bonding) at larger R 
values and negative (antibonding) at and below R e (table 7). The corresponding kinetic 
energy component T' 2s /2s is consistent with the nature of interference, negative at larger 
R values and positive at shorter R values. This orbital pair, therefore, has a binding 



density is constructed by 2p/2p orbital pair followed by that of B(2p)/N(2s) pair at all R. 
The corresponding kinetic energy components reveal that the interference interaction 
due to all these orbital pairs starts with a binding effect from a large R and grow steadily 
with decreasing #, and then increases at further closer approaches. The T' 2s i 2 p and 
T l 2 P i2s components pass through a minima at R e while the minima of the T' 2p/2p 
component is above R C} and increases very sharply at shorter R values indicating a 
stronger localisation of electron near the nuclei. However, starting from large R up to 
R e the major fraction of the interference energy is contributed by the 2p/2p orbital pair 
followed by that of B(2p)/N(2s) orbital pair, signifying that the B-N bond is principally 
formed due to electron sharing by 2p ff orbital of B with 2s and 2p a orbitals of N. 

Table 8 demonstrates that B(2s)/C(2s) orbital pair creates a constructive interference 
at longer R and a destructive interference at and below R e . The antibonding effect of 
this orbital pair is also noted in population analysis (table 3). The corresponding kinetic 
energy component T' 2s/2s is consistent with the nature of interference i.e. it is negative at 
larger values and positive at and below R e . The interference densities due to other 
orbital pairs increase monotonically with decrease in R and 2p orbitals, followed by 
B(2p)/C(2s) pair, construct the major fraction of interference density. The correspond- 
ing kinetic energy components, however, behave dissimilarly with those of borazane 
system. The T 2s/2p and T 2p!2s components decrease first, pass through minima, then 
increase with the decrease of R. The T 2p / 2s and T 2pj2p components decrease hand in 
hand at larger R values, but at shorter R values the latter increases very sharply and 
becomes positive. The sharper increase in the kinetic energy components of 2p orbital 
pair in this system may be due to the much closer approach of the interacting groups. 
Thus, the B-C bond is principally formed by 2 pa orbital of B, and 2s and 2 pa orbital of C. 
One more point to be discussed is the kinetic energy components of 2p orbital pairs 
below R e . These quantities are positive at shorter distances below R e in spite of the 
positive interference density of these orbital pairs (tables 7, 8). A closer look at the 
definition of kinetic interference energy (equations (16) and (17)) shows that the sign of 
this quantity depends on the signs of the element of bond order matrix and the integral 
T(AaBb >. The latter and not the former, should be very sensitive of I?. Although the 
atomic quantities T(Aa Aa) and T(Bb Bb] remain constant at all R because of non- 
inclusion of cluster promotion in the present calculation, T(AaBb) and S(AaBb) 
change with R to reverse the sign of the integral T(AaBb~) at shorter distances 
below R . 



4. Conclusion 

Thus we see that CNDO/2D calculation has reproduced the trend of charge distribution, 
in terms of orbital and overlap populations, of the ab initio method in H 3 B-NH 3 and 
H 3 B-CO fairly well. The CNoo/2 bond indices E AB and E% B , and CNDO/2-D derived 
quantities like P AB , p, p 1 and T' all indicate the binding situation as functions of 
internuclear separation. E AB and E AB are semi-empirical quantities while p is a very 
useful tool for studying various chemical phenomena (McWeeny 1954; Bamzai and 



the numerical results obtained were consistent with the concepts associated 
theoretical quantities. The CNDO/2-D method, therefore, appears to be a good a 
to extract useful information on chemical binding. 



References 

Ahlrichs R and Koch W 1978 Chem. Phys. Lett. 53 341 

Armstrong D R and Perkins P G 1969 J. Chem. Soc. (A) 1044 

Bach M C, Crasnier F, Labarre J F and Leibovici C 1973 J. Mol Struct. 16 89 

Bamzai A S and Deb B M 1981 Rev. Mod. Phys. 53 95 

Bethke G W and Wilson M K 1957 J. Chem. Phys. 26 1118 

Corre F 1981 J. Mol. Struct. 86 69 

Cowan R D 1949 J. Chem. Phys. 17 218 

Cowan R D 1950 J. Chem. Phys. 18 1101 

Das T P 1957 /. Chem. Phys. 27 1 

Datta R 1976 Indian J. Chem. A14 269 

Datta R and Datta M K 1978 Natl. Acad. Sci. Lett. 1 101 

Datta R and Datta M K 1978 Indian J. Chem. A16 616 

Datta R, Datta M K and Ghosh D C 1977 Indian J. Chem. 15 259 

Dill J D, Schleyer P V R and Pople J A 1975 J. Am. Chem. Soc. 97 3402 

Drissler F and Kutzelnigg W 1977 Theor. Chim. Acta 43 307 

Edmiston C and Ruedenberg K 1964 J. Phys. Chem. 68 1628 

Edmiston C and Ruedenberg K 1965 J. Chem. Phys. 43 S97 

Ehrenson S and Seltzer S 1971 Theor. Chim. Acta 20 17 

Ermler W C, Glasser F D and Kern C W 1976 J. Am. Chem. Soc. 98 3799 

Eyring H, Walter J and Kimball G 1944 Quantum chemistry (New York: John Wiley) 

Feinberg M J, Ruedenberg K and Mehler E L 1970 Adv. Quantum Chem. 5 27 

Fischer H and Kollmar H 1970 Theor. Chim. Acta. 16 163 

Frost A A 1970 Theor. Chim. Acta 18 156 

Fujimoto H, Kato S, Yamabe S and Fukui K 1974 J. Chem. Phys. 80 572 

Gordon M S and England W 1972 Chem. Phys. Lett 15 59 

Graham WAG and Stone F G A 1956 J. Inorg. Nuclear Chem. 3 164 

Ha T K 1976 J. Mol. Struct. 30 103 

Herzberg G 1966 Electronic spectra of polyatomic molecules (New York: Van Nostrand Reinhc 

Hoffman R 1964 J. Chem. Phys. 40 2474 

Hughes E W 1956 J. Am. Chem. Soc. 78 502 

Kato S, Fujimoto H, Yamabe S and Fukui K 1974 J. Am. Chem. Soc. 96 2024 

Labarre J F 1978 Struct. Bonding (Berlin) 35 1 

Layton Jr. E M and Ruedenberg K 1964 J. Phys. Chem. 68 1654 

Lippert E L and Lipscomb W N 1956 J. Am. Chem. Soc. 78 503 

Lloyd D R and Lynaugh N 1970 Chem. Commun. 1545 

Lloyd D R and Lynaugh N 1972 J. Chem. Soc. 68 947 

Lowdin P O 1950 J. Chem. Phys. 18 365 

Mclver Jr J W, Coppens P and Nowak D 1971 Chem. Phys. Lett. 11 82 

McWeeny R 1951 J. Chem. Phys. 19 1614 

McWeeny R 1954 Proc. R. Soc. A223 63 

Moireau M Cl and Veillard A 1968 Theor. Chim. Acta 11 344 

Mulliken R S 1955 J. Chem. Phys. 23 1833 

Nakatsuji H 1974 J. Am. Chem. Soc. 96 24, 30 

Palke W E 1972 J. Chem. Phys. 56 5308 



Purcell K F and Martin R L 1974 Theor. Chim. Acta 35 141 

Ransil B J 1960 Rev. Mod. Phys. 32 245 

Redmon L T, Purvis III G D and Bartlett R J 1979 J. Am. Chem. Soc. 101 2856 

Roothaan C C J 1951 J. Chem. Phys. 19 1455 

Rue R R and Ruedenberg K 1964 J. Phys. Chem. 68 1976 

Ruedenberg K 1962 Rev. Mod. Phys. 34 326 

Runtz G R and Bader R F W 1975 Mol. Phys. 30 129 

Shillady D D, Billingsley II F P and Bloor J E 1971 Theor. Chim. Acta 21 1 

Shore S G and Parry R W 1955 J. Am. Chem. Soc. 77 6084 

Slater J C 1933 J. Chem. Phys. 1 687 

Sundaram S and Cleverland F F 1960 J. Chem. Phys. 32 166 

Taylor R C 1957 J. Chem. Phys. 28 1131 

Umeyama H and Morokuma K 1976 J. Am. Chem. Soc. 98 7208 

Veillard A, Levy B, Daudel R and Gallais F 1967 Theor. Chim. Acta 8 312 

Wilson Jr. C W and Goddard III W A 1970 Chem. Phys. Lett. 5 45 

Wilson Jr. C W and Goddard III W A 1972 Theor. Chem. Acta 26 195 

Zirz C and Ahlrichs 1981 J. Chem. Phys. 75 4980 



Reaction of the carbonate radical with substituted anilines 

T P ELANGO, V RAMAKRISHNAN, S VANCHEESAN and 
J C KURIACOSE* 

Department of Chemistry, Indian Institute of Technology, Madras 600036, India 

MS received 3 February 1983; revised 12 May 1983 

Abstract. Rate constants for the reaction of carbonate radical with aniline and some para- 
substituted anilines have been determined by the flash photolysis technique. Using ff+para 
values the rate constants at pH 8-5 correlate very well with the Hammett equation yielding 
p - 1. The carbonate radical oxidises aniline giving the anilino radical. The products so 
formed have been identified through studies under conditions of continuous irradiation. 

Keywords. Flash photolysis; carbonate radical; rate constants; substituted anilines. 



1. Introduction 

The carbonate radical (E ed = 2-1 V) (Endicott 1975) can react with aromatic substrates 
through direct electron transfer or through an electrophilic addition or substitution 
(Ross and Neta 1979; Chen et al 1975). The reaction of the carbonate radical with a 
series of para-substituted anilines has been examined with a view to obtain indications 
about the mechanism. 

2. Experimental 

2.1 Materials 

The complex [Co(NH 3 )4CO 3 ]ClO4 is prepared according to literature procedure 
(Rochow 1960; Chen et al 1973). All the anilines were purified by recrystallisation. In the 
case of liquid substances, freshly-distilled samples were used for preparing solutions. 
For flash photolysis experiments the pH of the solutions was adjusted using KH 2 PO 4 
and NaOH and for continuous photolysis studies NaOH alone was used for adjusting 
the pH. 

2.2 Measurements 

Flash photolysis experiments were carried out using a 10cm quartz cell in a Nortech 
flash photolysis unit type FPX-! and the transient was recorded using data display 
system type DFR-2 (Nortech Laboratories, England) combined with a chart recorder 
(Siemens Kompensograph 111). The flash lamps dissipated up to 200 J of energy with a 
half peak duration of about 30jusec. The decay of the carbonate radical was monitored 
by following the absorbance at 600 nm and measurement made at 23 1C. All 



solutions were prepared with triply-distilled water and deaerated with purified N 2 
before flashing. However, it was found that dissolved oxygen has no effect on the kinetic 
data. Only freshly-prepared solutions were used for flash photolysis studies to exclude 
possible thermal reaction and were discarded after a single flash. 

2.3 Product analysis 

Product analysis studies were carried out by continuous photolysis with a 48 W low 
pressure Hg vapour lamp (Rayonet RUL 2537 A). Before the carbonatotetramine- 
cobalt(III) salt is used as a source of carbonate radical for continuous irradiation 
studies, experiments were conducted leaving the reaction mixture in the dark and it was 
noticed that in dark this Co(III) complex oxidises aniline slowly. In order to further 
minimise their reaction, the solution is well cooled and the complex salt is added at the 
last moment before irradiation. After irradiation the products are immediately 
extracted with ether. In a typical experiment, a solution containing 10" 2 M complex 
and 10" 3 M aniline at pH 8-5 is irradiated for 30min after deaeration with purified N 2 
or Ar. The residue from the ether extract is separated and identified by TLC. 4- 
aminodiphenylamine, azobenzene and benzidine are the major products and 2- 
aminodiphenylamine and hydrazobenzene are formed in trace amounts. In addition, 
phenylhydroxylamine is also formed. Azobenzene is formed by the rapid oxidation of 
hydrazobenzene. Therefore in these experiments azobenzene was identified as a major 
product and hydrazobenzene as a minor product, uv spectrum of these separated 
products compared well with that of authentic samples. The separated amines are also 
mixed with p-dimethylaminobenzaldehyde and the visible spectra of the resulting 
derivatives (Zechner et al 1976) compared well with those of authentic samples. Under 
these conditions aniline itself without added complex underwent photolysis yielding 
the same products, except phenylhydroxylamine, but in considerably reduced yields. In 
mixtures the Co(III) complex is the major absorber of the radiation and not aniline. At 
254 nm, the extinction coefficients for cobalt(III) complex and aniline are found to be 
1 1340 and 669 M ~ i cm ~ 1 respectively. In the dark the complex plus aniline gets slowly 
oxidised at 40^ C to give phenylhydroxylamine and the products such as 4-aminodi- 
phenylamine, hydrazobenzene, etc which were isolated in photolytic conditions were 
conspicuously absent. These results show that the greater yield of 4-aminodiphenyl- 
amine, azobenzene, benzidine and 2-aminodiphenylamine that are obtained on 
irradiating the carbonate tetraminecobalt(III) complex with aniline result from the 
reaction of the carbonate radical with aniline. 

3. Results 

The carbonate ion radical is produced on flashing a solution containing about 
2 x 10" 5 M [Co(NH 3 ) 4 CO 3 ]ClO4 by charge transfer to metal (CTTM) transition (Cope 
and Hoffman 1972; Chen et al 1973). 



order in the absence of any scavenger (decay constant 2 x 10 7 M 1 sec *) and becomes 
pseudo-first order in the presence of added scavenger. This pseudo-first order rate 
constant is determined at least three different initial concentrations of scavenger and at 
each concentration at least four kinetic curves were processed. Second order rate 
constants were determined from the slope of the plots of the pseudo-first order rate 
constants vs [scavenger] (figure 1). The rate constants obtained for aniline and some 
parasubstituted anilines at pH 8-5 are given in table 1. These data are subject to an error 
inherent in flash photolysis studies (+ 10%). 

The effect of pH on the rate constant for the reaction of carbonate ion radical with 
aniline is studied. The rate constant remains almost constant in the pH range 6-11. 




^ 1-0 
x 10 D M [5] 



1.5 



Figure 1. Dependence of /c obs on scavenger concentration at pH = 8-5, a. aniline 
b. p-aminobenzoic acid c. p-nitroaniline. 



Table 1. Rate constants for the reaction of CO 3 H'with aniline and p-substituted anilines 
(pH = 8-5). 



Scavenger 



Aniline 

p-chloroaniline 

p-bromoaniline 

p-aminobenzoic acid 

p-aminoethylbenzoate 

p-nitroaniline 



M 



5-0 x 10 8 
4-3 x 10 8 
3-8 x 10 8 
2-0 x 10 s 
2-0 x 10 s 
7-3 x 10 7 



independence of the rate constant shows that the intrinsic reactivity of CO 3 H and 
CO J towards aniline is the same. 

4. Discussion 

To understand the effect of substituent on the reactivity of aniline with carbonate 
radical the rate constants given in table 1 are correlated with appropriate Hammett 
substitution constants. With a* values (Brown and Okamoto 1958) a good 
correlation is obtained (figure 2). The p value of 1 obtained is similar to that observed 
for the reaction of carbonate ion radical with substituted phenols (Moore et al 1977). 
Crable and Kearns (1962) found that the ionization potentials of the 
p-substituted anilines correlate well with a* values. Thus it is possible that the rate 
constants are influenced by the ionization potentials of the amine suggesting possible 
electron transfer from aniline to the carbonate radical. Attempts were therefore made 
to observe the possible aniline-radical intermediate. Since the pK a of the anilino radical 
C 6 H 5 NH2 is 7 (Land and Porter 1963), at pH 8-5 it deprotonates to give C 6 H 5 NH'. 
For the protonated and basic forms of the anilino radical, absorption maxima of 423 
and 300 nm respectively have been reported by Land and Porter (1963). When a 
solution containing 3xlO" 5 M [CO(NH 3 )4CO 3 ]C1O4 and 2xlO~ 5 M aniline at 
pH 8-5 is flashed with about 125 J of energy there is very little absorption in the 423 nm 
region and the absorption around 300 nm could not be followed due to limitations of 



-0-8 - 




-COO 
po-COOEt 



-CU 



0-A 



0-8 



'para 



the smaller flash energy. Since the pK a value of the bicarbonate radical is 9-6 0-3 at 
both pH 6 and 8-5 the bicarbonate radical will exist mainly in the protonated form 
CO 3 H'. Hence at pH 8-5 also the formation of C 6 H 5 NH2 is expected but will 
immediately deprotonate to give C 6 H 5 NH' which dimerises to give hydrazobenzene 
and other observed products (scheme 1). 

Though the <jp values correlate very well with IP of substituted aniline, figure 2 
cannot be taken as a definite indication of the mechanism of the reaction, i.e. electron 
transfer or radical attachment. The pa* correlation has also been applied to 
electrophilic aromatic substitution (Exner 1972) and step 1 of scheme 1 may involve the 
intermediacy of an adduct (scheme 2). 

The adduct could rearrange by an overall electron transfer mechanism to give the 
anilino radical. Since no specific transient other than anilino radical could be observed 
in the region 350-650 nm, the formation and collapse of the adduct may be very rapid 
and these two processes may be completed before any observation could be made. It is 
believed that such an adduct may indeed exist since a similar reactivity for both the 
forms of the radical towards aniline has been observed. If a direct electron transfer 
mechanism were operating, then one would expect the uncharged protonated 'form 
CO 3 H'to be a stronger oxidising agent and thus to react more rapidly in an electron 
transfer process than COJ. However both forms could show similar reactivity if H 
abstraction or addition modes were operative. 



SCHEME - 1 



C0 3 H '+ C G H 5 NH 2 "" C 6 H 5 NH 2 * C 

C 6 H 5 NH 2 *- C 6 H S NH + H* (>pH 7) 




azobenzene 




\=/ 




NH, 



SCHEME -2 

NH 2 NH 2 

-C0 3 H - fil + C0 3 H" 



aniline to give both cyclohexadienyl type radical ana anmno radical, i ne less reac 
CO 3 H" radical is more selective and gives only anilino radical. 

The reaction of carbonate radical with aliphatic amines is also studied and the 
constants are found to be two or three orders lower than those for aniline. These 
constants do not give a satisfactory correlation with their ionization potentials. V 
amines like KCH 2 NH 2 products like KCHO are identified, probably formed 
a-hydrogen abstraction. Detailed studies of these systems are in progress. 

5. Conclusion 

It may be concluded that carbonate radical oxidises aniline to anilino radical. Anil 
containing electron donating p-substituents have higher rate constants at pH 
compared to unsubstituted aniline whereas the converse is observed for anil 
containing electron withdrawing p-substituents. 

Acknowledgement 

Dr B Viswanathan's assistance in the operation of the flash photolysis instrumei 
gratefully acknowledged. One of the authors (TPE) is grateful to the uoc, New Delhi 
Government of Tamil Nadu for providing a teacher fellowship. 

References 

Brown H C and Okamoto Y 1958 J. Am. Chem. Soc. 80 331 

Chen S N, Cope V W and Hoffman N Z 1973 J. Phys. Chem. 77 1111 

Chen S N, Hoffman M Z and Parsons G H Jr 1975 J. Phys. Chem. 19 1911 

Christensen H 1972 Int. J. Radial. Phys. Chem. 4 311 

Cope V W and Hoffman M Z 1972 Chem. Commun. 227 

Crable G F and Reams G L 1962 J. Phys. Chem. 66 436 

Endicott J F 1975 in Concepts of inorganic photochemistry (eds) A W Adamson and P D Fleischauer ( 

York: Wiley Interscience) 
Exner O 1972 in Advances in linear free energy relationships (eds) N B Chapman and J Shorter (New ^ 

Plenum Press) 

Land E J and Porter G 1963 Trans. Faraday Soc. 59 2027 

Moore J S, Philips G O and Sosnowski A 1977 Int. J. Radial. Biol. Relat. Stud. Phys. Chem. Med. 31 
Rochow E G ed 1960 in Inorganic synthesis (New York: McGraw-Hill) Vol. 6 p. 173 
Ross A B and Neta P 1979 Natl. Stand. Ref. Data Ser. (U.S. Natl. Bur. Stand.) No. 65 
Zechner J, Prangova Lilia St, Grabner G I and GetoffN 1976 Z. Phys. Chem. Neue Folge 102 137 



Boron complexes of sulphur containing Schiff bases derived by the 
condensation of S-methyl or S-benzyl dithiocarbazate with /?-diketones 

P K SINGH and J P TANDON* 

Department of Chemistry, University of Rajasthan, Jaipur 302004, India. 

MS received 13 January 1983; revised 3 May 1983 

Abstract. Tetra-coordinated boron derivatives, (EtO)B(DTZ) and (DTZH)B(DTZ), (where 
DTZ" " and DTZ~ represent the anions of the Schiff base DTZH 2 ) have been synthesized by 
1 : 1 and 1 : 2 molar reactions of triethoxyborane with bibasic tridentate Schiff bases, derived by 
the equimolar condensation of S-methyl or S-benzyldithiocarbazate with acetyl acetone or 
benzoyl acetone. Further 1 : 1 derivatives have been shown to undergo replacement reactions 
with t-butyl alcohol, showing thereby the labile nature of the ethoxy group. Based on infrared 
and proton magnetic resonance spectral studies and monomeric nature, suitable structures 
have been assigned to these derivatives. 

Keywords. Boron complexes; dithiocarbazate derivatives; Schiff bases; 'H NMR spectra; 
IR spectra. 



1. Introduction 

Extensive studies on the transition metal complexes of dithiocarbazate Schiff bases 
have been carried out and a review featuring their geometry and configurations has also 
appeared (Ali and Livingstone 1974). However, similar investigations concerning the 
non-transition element complexes are scanty (Pardhy et al 1980; Agarwal et al 1980). In 
earlier communications (Singh et al 1980; Singh and Tandon 1977) a variety of boron 
derivatives with nitrogen donor ligands have been reported and some of these have 
been found to possess considerable biological activity (Singh and Tandon (un- 
published)). Recently, the physiological aspects of boron complexes have also been 
reviewed (Niedenzeu 1979). 

Some new boron derivatives having B-S bond, are synthesized by the reactions of 
triethoxyborane with the Schiff bases having the donor system HSNOH and which 
may be structurally represented as follows: 



HC 



H tf- 



(Thione form) (Thiolo form) 

I, R = -CH 3 , R' = -CH 3 (AcAcMTZH 2 ) 

II, R = -CH 3 , R 1 = -CH 2 C 6 H 5 (AcAcBTZH 2 ) 

III, R = -C 6 H 5 , R' = -CH 3 (BzAcMTZH 2 ) 

IV, R = -C 6 H 5 , R' = -CH 2 C 6 H 5 (BzAcBTZH 2 



The desired products are obtained by 1 : 1 and 1 : 2 molar reactions of triethoxyborane 
with the ligands AcAcMTzH 2 , AcAcBTzH 2 , BzAcMTzH 2 , BzAcBTzH 2 . However, a 
2:3 molar reaction simply yielded a mixture of the products formed by 1:1 and 1:2 
molar reactions. The 1 : 1 product could be isolated using a mixture of benzene and 
p-xylene. These reactions were quite facile due to the weaker B-OR bonding as well as 
the reactive nature of the dithiocarbazate Schiff bases (Ali and Livingstone 1974). 



B(OEt) 3 +HO N SH^EtO-B N + 2EtOH 

^~\ s~\ s~ o 

B(OEt) 3 + 2HO N SH-+N 



(where HO N SH represents the donor system of bibasic tridentate Schiff base). 

The 1 : 1 products further undergo replacement reactions with t-butyl alcohol as 
indicated below: 

ff 0OEt 



^- in excess ^- 

The resulting products are obtained as coloured, nonvolatile and highly viscous 
liquids or solids. These are soluble in common organic solvents, monomers and 
unstable in the open atmosphere probably due to the weaker B-S bond. 

2.1 IR spectra 

The IR spectra of the SchifF bases in solid form display strong bands at 1 500 and 
1050cm" 1 assignable to v c _ N and v c=s respectively Sahni and Kapoor (1979). These 
bands support the presence of the thione form; however, in solution spectra in 
ethanol, both these bands disappear and a sharp band at ~ 2570cm" 1 is observed 

due to SH which is indicative of the thiol form. These studies probably indicate that 

i 
the -NH-C=S group exists mainly in the solid form, whereas in solution an equilibrium 

exists between the two tautomeric forms. 

A strong band appears in the region, 3200-3125 cm' 1 due to the hydrogen bonded 
V OH /V NH which is not observed in the 1 : 1 boron derivatives, due to the deprotonation of 
OH/NH group and the chelation of oxygen to boron. However, in the spectra of the 
1 : 2 derivatives the characteristic band of NH is observed indicating that one of the 
functional groups of at least one Schiff base does not undergo deprotonation. 

The shifting of the frequency due to azomethine stretch v c=N to the higher side, 
( ~ 1600 cm" ^-163^ cm" l ) in the boron derivatives is probably due to an increase in 
the C=N bond order after the formation of N -> B bond (Samuel et al 1970). All these 
derivatives show a strong band at ~ 1540cm" 1 which may be due to the VB_ N as 
reported earlier (Wang et al 1970). The v,^ (asym) appears at ~ 1310cm" 1 in all the 
boron derivatives (Singh et al 1980). 

2.2 NMR spectra 



Compound 



a b c d e f 

(a') (b') (c') (d 1 ) (e') 



H 3 C \ 
c - 



C - 



. 
d C 



i w e 

n. n-o 
N N4=i SCH 



2-10 2-40 2-56 5-83 10-75 



HC. 



,OCHCH 



23 
9 

S 
SCH 



2-21 2-94 2-61 6-28 3-53 1-13 



H 3 C S 

/~\/ ~" C 
H ri C v / S 



d s 









C a H 3 S C-S 
SCH, 



2-38 2-92 2-57 6-13 10-52 



(6) ppm for the various protons are listed in table 1 and the following points appear to 
be significant from the structural point of view: (i) In 1 : 1 derivative, the disappearance 
of hydroxyl (d 10-75 ppm) and thiolo proton signal (S 445 ppm) shows chelation of 
boron through both oxygen and sulphur atoms, whereas in 1 : 2 derivative the 
appearance of the thiolo proton signal at 6 4.25 ppm, clearly indicates that at least in 
one of the ligand moieties one functional group remains unbonded, (ii) In 1 : 1 
derivative, the methine proton and one of the methyl proton (labelled d and b 
respectively in table 1) signals show downfield shifting leaving behind two methyl 
proton (labelled a and c table 1) signals almost unchanged. This may be due to the 
boron acquiring a tetracoordinated state after accepting a lone pair of electrons from 
the azomethine nitrogen, (iii) In 1 : 2 derivative, however, two sets of proton signals are 
observed for every proton. One set remains at almost the same position as in the ligand, 
thereby showing that one of the ligands is acting as tridentate and another one as 
unidentate as shown in table 1. 



3. Experimental 

A glass apparatus fitted with quickfit interchangeable joints was used and the reactions 



'55 

a 



<=> -r <? -r* 



.a 



N 



Compound 



o-N 
<'5b >><: 




8 



Triethoxyborane was prepared by refluxing boric acid (S. Merck) with absolute ethanol 
in dry benzene (BDH) and removing water azeotropically (Lippincott). It was distilled 
before use and analysed: Found: B, 7-44; OC 2 H 5 , 91-97 % Calcd. for B(OEt) 3 : B, 7-41; 
OC 2 H 5 , 92-59%. 

S-methyldithiocarbazate and S-benzyldithiccarbazate were prepared as reported 
earlier (Ali et al 1971; Ali and Bose 1977). To prepare the Schiff bases, a weighed amount 
of S-methyl- or S-benzyl dithiocarbazate was dissolved in ethanol and mixed with an 
equimolar amount of j?-diketone (ethanolic solution). The reaction mixture was kept 
over a water bath for about 10 min and left overnight at room temperature. The crystals 
which separated out were filtered off and recrystallized from the same solvent. 

3.2 Analytical methods and physical measurements 

Boron was estimated as reported earlier (Singh and Tandon 1977). Nitrogen was 
determined by the KjeldahPs method and the sulphur was analysed by the literature 
method (Saraswat et al 1977). Carbon and hydrogen were analysed by the carbon 
hydrogen analyzer (Coleman-5602). Ethanol liberated in the reactions was estimated 
oxidimetrically (Bradley et al 1950). Molecular weight was determined ebullio- 
scopically in boiling benzene using thermistor sensing. 

IR spectra were recorded in Nujol mulls on a Perkin Elmer 577 grating infrared 
spectrophotometer and PMR spectra were scanned in carbon tetrachloride (using TMS as 
internal standard) on a Perkin-Elmer RB-12 spectrometer. 

3.3 Synthesis of boron Schiff base complexes 

A weighed amount of triethoxyborane was mixed with the calculated amount of Schiff 
base in dry benzene ( ~ 75 ml). The reaction mixture was refluxed over a fractionating 
column and the ethanol liberated was collected azeotropically with benzene. The 
completion of the reaction was ascertained by the ethanol estimation. After its complete 
removal, the excess of the solvent was distilled off and the products dried under reduced 
pressure (2 mm) for 2 hr. Further details are listed in table 2. 

3.4 Exchange reactions of boron Schiff base derivatives 

f-Butanol in excess was mixed with dry benzene solution of ethoxyboron Schiff base 
derivatives and the reaction mixture was refluxed over a ratio head. The ethanol was 
collected azeotropically with benzene and the rest of the procedure was the same as 
described in 3.3. 

Acknowledgement 

One of the authors (PKS) is grateful to the CSIR, New Delhi for the award of a fellowship. 

References 



T xi 1 nor c. 



All M A, Livingstone S E and Phillips D J 1971 Inorg. Chem. Act. 5 119 

Bradley D C, Halim F M and Warlaw A 1950 J. Chem. Soc. p. 3450 

Lippincott S B, U S Pat 2 642 453 (1953 to Standard Oil Development Co.) 

Niedenzu K 1979 J. Organometal. Chem. 180 89 

Pardhy S A, Gopinathan S and Gopinathan C 1980 Indian. J. Chem. A19 130 

Sahni S K and Kapoor R N 1979 Indian. J. Chem. A18 456 

Samuel B, Snaith R, Summerford C and Wade K 1970 J. Chem. Soc. 2019 

Saraswat B S, Srivastava G and Mehrotra R C 1977 J. Organometal. Chem. 137 301 

Singh P K, Singh H B and Tandon J P 1980 Synth. React. Inorg. Met-Org. Chem. 10 443 

Singh H B and Tandon J P 1977 Synth. React. Inorg. Met-Org. Chem. 1 547 

Singh P K and Tandon J P Unpublished results 

Wang T T, Bausse P J and Niedenzu K 1970 Inorg. Chem. 9 2150 



Reduction mechanism of I,4-diamino-2,3-anthraquinone 
disulphonic acid 

F CAPITAN*, E ALVAREZ-MANZANEDA and J L VILCHEZ 

Department of Analytical Chemistry, Analytical Section of C.S.I.C., Faculty of Sciences, 
Granada, Spain. 

MS received 22 February 1983; revised 13 July 1983 

Abstract. The potentiometric acid constants of l,4-diamino-2,3-anthraquinonedisulphonic 
acid were calculated. The sequence of dissociation is discussed. The reduction of the reagent at 
a dropping mercury electrode has been investigated. The polarograms of reagent show two 
waves, whose adsorption and diffusion nature is respectively established. The reaction orders, 
together with Tafel's slopes have been calculated. 

Keywords. Anthraquinones; polarography; potentiometry; reduction mechanism 



1. Introduction 

The diaminoanthraquinones are selective and sensitive reagents suitable for spectro- 
photometric and fluorimetric determination of a good number of metallic ions 
(Krausz et al 1963), but no attention seems to have been paid yet to their electro- 
chemistry. l,4-diamino-2,3-anthraquinonedisulphonic acid has been previously 
studied with respect to some analytical applications and has been proposed for the 
spectrophotometric determination of Au(III) (Capitan et al 1977), Pd(II) (Capitan et al 
1982) and as an acid-base indicator (Capital et al 1978). 

In this paper the potentiometric dissociation constants are determined and the 
reduction of this reagent on a mercury electrode is studied. 

2. Experimental 

l,4-diamino-2,3-anthraquinonedisulphonic acid was synthesized according to 
Pattison's patent (Pattison et al 1957) by condensing l,4-diamino-2,3-dichloroanthra- 
quinone with boric acid and sulphuric acid to form a boric acid ester complex. The 
excess acid is neutralized and sodium sulfite added, replacing thus the chlorine atoms 
with sulphonic groups when heated at about 95-100C. The reagent was identified by 
elemental analysis, IR and NMR. 

The acid-dissociation constants of reagent have been determined by the poten- 
tiometric Bjerrum's (1941) method at ionic strength n 0-1, from the titrations of the 
free acid, monosodium salt and disodium salt performed with a Metrhom-Herisau 
E-536 potentiograph. 

The i-E curves were registered either automatically or traced point by point using a 



60 



F Capitan, E Alvarez- Manzaneda and J L Vilchez 



471 Amel multipurpose unit with the damping circuit completely suppressed. The cyclic 
voltammetry curves were recorded on a Tektronix SSP 3 polarograph. The potentials 
were measured vs. SCE with a Radiometer PHM-84 potentiometer which was also used as 
a pH-meter. 

The polarographic measurements were made using a thermostated Amel 494 cell. A 
saturated calomel reference electrode was used. The working electrode was a mercury 
capillary with the following characteristics: rate of mercury flow m = 1-95 mg/sec, drop 
time t 5-15 sec, open circuit, in a buffered solution at pH 12-15 and the height of the 
mercury column h 42-50 cm. The cyclic voltammetry curves were found out using a 
hanging drop electrode as working electrode. 

All the reagents used were Merck AR grade. As a supporting electrolyte solution, a 
buffered solution of 6-66 x 10~ 2 MNa 3 PO 4 and 0-1M. H 3 PO 4 mixed in varying 
proportions according to the pH desired was used. The ionic strength was adjusted with 
KNO 3 to 0-24 M. All the measurements were done in an atmosphere of nitrogen and at 
a temperature 25-0 + 0-1 C. 

3. Results 

The acid-dissociation constants of l,4-diamino-2,3-anthraquinonedisulphonic acid 
were determined potentiometrically (Bjerrum 1941) at ionic strength /i = 0-1. A 
maximum of two dissociation constants have been determined in the range of pH 
investigated and their pK values are pK t = 3-91 and pK 2 = 8-15. 

The l,4-diamino-2,3-anthraquinonedisulphonic acid reduced on the DME produce 
two polarographic waves from aqueous solutions at concentrations smaller than 
l-6x!0~ 3 M, Both waves were always observed throughout the complete pH 
range studied (2-00-12-71). Figure 1 shows a representative polarogram at pH 12-50 
= - 0-616 V, and E = - 0-848 V. 



1.00 - 



0.75 



0.50 



0.25 



current on me neignt or me mercury column ana me temperature. 1 nus tne aasorpt 
and diffusive nature of the first and the second wave respectively were proved (Zum 
1969). 

The dependence of the half- wave potential on pH was studied as well, to determi 
how many hydrogen ions are involved in the reduction. Polarograms were recorded 1 
2 x 10""* M solutions in the pH range 5-60-12-72. EL and n a values were calculated 
each case. Figure 2 shows the plots of 1 vs pH. One can observe two straight lines t 
slopes of which are 62 mV and 28 mV depending on whether pH is lower or higher th 
8-27. Accordingly, the number of hydrogen ions involved in the reduction must 
either p = 2 or p = 1 respectively. 

On the other hand, we checked whether the limiting currents are proportional to t 
concentration of the l,4-diamino-2,3-anthraquinonedisulphonic acid in the bulk oft 
solution, in the range 5 x 10~ 3 M-4 x 10 ~ 5 M at pH 12-50. The values of the slo 
of the linear segments are: l-65juAL/mM and 3-10juAL/mM for the first and t 
second wave respectively. Hence, the diffusion current constants are / = i d /m zl ^t 1 
C = 607 D 1 /2 n 1-51 calculated for the second wave, and the diffusion coefficient is 
= 1-55 x 10~ 6 cm 2 /sec. Such / and D values agree with those previously published f 
some anthraquinone (Capitan et al 1979, 1980, 1981). 

The half-wave potentials , remains independent with an increase in concentratic 

The transfer coefficient determined from Tafel's slope is a = 0-48 and the order 
reaction with respect to the concentration is: 

|(dlogi/dlogC)| =l 

calculated from the second wave. 

We carried out controlled potential microcoulometry to determine n, i.e., t 
number of electrons involved in the reduction. It appears that the reduction 



0,9 



0,8 



0,7 



0,6 




10 



concentration were registered showing always two cathodic peaks and one anodic peak 
respectively. The main observed features are: 

(a) Successive voltammograms obtained at different sweep rates show that the first 
wave disappears when a second cycle is imposed immediately after the first and the 
sweep rate is higher than 20mV/sec (figure 3). This confirms the adsorption nature of 
the first wave. Such behaviour has been observed by Ali-Qureshi et al (1979) for a 
similar reagent. 

(b) The cathodic (positive) and anodic (negative) peak currents for the second wave are 
virtually identical; the peak separation E P(C) E P(U] being 33 mV, which is not far from 
the theoretical value of 30 mV for a bielectronic transference. Hence, one can say that 
the reduction of reagent is reversible under such conditions. On the other hand the plots 
of i p vs. log v at different concentrations are linear with slopes of 0-5, in agreement with 
theoretical value. 

(c) The a and reactions order values calculated from this technique agree with those 
previously calculated polarographically. 

4. Discussion 

The acid dissociation constant values obtained potentiometrically as well as the strong 
acidic nature of the sulphonic groups (being thus extensively dissociated) suggest that in 



20 - 




-10 - 



-0.600 -0.800 -1.000 

E(v) vs see 



-1.200 



(i) 

( } 



pH<8.20 
(2) 



M T + " J ~ M I ~ ^ II I 

NH 3 NH 3 NH 2 

(a) (b) (c) 




(3) 



pH>8.20 NH 2 OH NH? 

,. , t- | , ^ 

+ . r^^N^V 

- * 2 e- + H == 





aqueous solution the l,4-diamino-2,3-anthraquinonedisulphonic acid is structured 
mainly as in scheme la. Owing to its structure, (a) is a rather strong acid (pK^ = 3-91) 
and it may be easily dissociated to give the structure as in scheme Ib. Finally it can 
dissociate (pK 2 = 8-15) to give a structure as in scheme Ic. 

These dissociation steps agree with those found for similar molecules (Garcia- 
Sanchez et al 1970). 

Moreover, here, the polarographic curves show that the reduction of the 
l,4-diamino-2,3-anthraquinonedisulphonic acid proceeds by one bielectronic step 
diffusion controlled in basic medium, showing specific adsorption of the reagent at the 
mercury electrode. The reagent requires two electrons and two protons at pH lower 
than 8-20 and two electrons and one proton at pH higher than 8-20, to produce the 
respective hydroquinone. The shape of the cyclic voltammetry curves is, in agreement 
with the diagnostic criteria proposed for a EE mechanism at the investigated scan rates. 

From the results and conclusions obtained and in accordance with the dissociation 
steps of the reagent, we propose for the reduction of l,4-diamino-2,3-anthraquinonedi- 
sulphonic acid, a mechanism with the following reaction pathways: (a) At pH lower 
than 8-20 the predominant form is as in scheme Ib, which captures two electrons and 
two protons to form a hydroquinone system, (scheme 2). (b) At pH higher than 8-20 the 
predominant form is as in scheme Ic, which captures two electrons and one proton to 
form a hydroquinone system (scheme 3). 



References 

Ali-Qureshi G, Suehla G and Leonard M A 1979 Analyst 104 705 

Bjerrum ] 1941 Metal amine formation in aqueous solution (Copenhagen: Haase P) 



,.,J 1/4 AH"! 



Capitan r, Guiraum A and vilcnez J L iyl Can. J. Lnem. 5y 1ZU1 

Capitan F, Garcia-Sanchez F and Gomez-Hens A 1982 Univ. Ind. Santander. Bucaramanga, Col 

Garcia-Sanchez F, Bosch F and Estela J M 1970 An. Quim. B76 273 

Krausz I, Endroi H and Havas P 1963 Magy. Kern. Foly. 69 519 

Pattison D B 1957 Du Pont de Nemours and Company U S Patent Office 2 795 593 June 

Zuman P 1969 The elucidation of organic electrode processes (New York: Academic Press) p 



Spectral, magnetic and thermal studies of transition metal complexes 
of <5(3-carboxy, 4-hydroxy benzoyi) pentanoic acid 

M N PATEL* and M R CHAUDHARI 

Chemistry Department, Sardar Patel University, Vallabh Vidyanagar 388 120, India 
MS received 30 August 1982; revised 24 December 1982 

Abstract. Cu(II), Ni(II), Co(II), Mn(II) and Zn(II) form 1 : 2 complexes while Fe(III) forms 
1 : 3 complex with i5(3-carboxy, 4-hydroxy benzoyi) pentanoic acid. Their structures have been 
proposed on the basis of analytical, spectral, thermal and magnetic measurements. The 
infrared spectra provide an evidence for the replacement of the aromatic carboxylic group 
hydrogen but not the phenolic group hydrogen. 

Keywords. 5 (3-carboxy, 4-hydroxy benzoyi) pentanoic acid; salicylic acid; transition metal 
complexes 

1. Introduction 

In recent years metal carboxylates have attracted considerable attention in chemistry 
because of their importance in industry and their interesting structure (Bassi et al 1980). 
Inoue et al (1964) reported magnetic moment of Cu(II) salicylate. Koppikar and 
Soundararajan (1976) studied diiodosalicylates of rare earths. A large number of 
derivatives of salicylic acid with metal halides have been reported (Goyal and Khosla 
1980). However no study of transition metal complexes of <5(3-carboxy, 4-hydroxy 
benzoyi) pentanoic acid has been reported so far. Hence this study. 



(M = Cu + 2 , Co + 2 , Ni + 2 , Mn + 2 , and Zn + 




2. Experimental 

2.1 Preparation of pentanoic acid 

Polyphosphoric acid (50 g) was added to a mixture of salicylic acid (0-05 mol, 6-9 g) and 
adipic acid (0.075 mol, 10.95 g). After mixing, the reaction mixture was heated at 1 10C 
for 4 hr. The solid obtained was added to the ice-cold water, filtered and washed with 
hot water. It was purified by dissolving in 10% alkali solution and precipitated as a 
brownish green coloured compound with ice cold 10 % HC1. It is soluble in alcohol with 
a melting point of 150C. 



2.2 Preparation of complexes 

The complexes of Cu(II), Ni(II), Co(II), Mn(II), Zn(II) and Fe(III) were prepared by 
mixing the solution of metal chloride in water and solution of <5(cHB) PA in absolute 
alcohol in a stoichiometric 1 : 2 molar ratio (metal : ligand) for bivalent metal ions and 
1 : 3 for Fe + 3 . Precipitates were obtained by addition of sodium acetate. The complexes 
were washed with hot water and ethanol. 

2.3 Elemental analyses 

Carbon and hydrogen were microanalysed by Coleman carbon hydrogen analyzer. 

2.4 Metal content in the complexes 

Known amounts of the metal complexes were decomposed with AR cone. HC1, HNO 3 , 
HC1O 4 and H 2 SO 4 . The residue was cooled, dissolved in water and made up to a 
known volume in a volumetric flask. All the metal ions were determined by EDTA 
titration using appropriate indicator. 

2.5 Physical measurements 

Magnetic susceptibility of the complexes at room temperature was determined by 
Gouy method using Hg[Co(NCs) 4 ] as the calibrant. Molar susceptibilities were 
corrected for diamagnetism of the constituent elements using Pascal's constants. The IR 
spectra of ligand and complexes were recorded in the form of KBr pellet on UR-10 
spectrophotometer in the range 3600-400 cm~ *. Diffuse reflectance spectra in the solid 
state were obtained from a Beckman DU spectrophotometer using MgO as reference. TG 
data were obtained from an assembly capable of producing temperatures up to 800C 
with a heating rate 10 min" 1 . 



3. Results and discussion 

The important thermal, spectral and magnetic moment data of the complexes are 
summarised in table 1. The infrared spectra provide important information about the 
attachment of ligand to the metal ions. IR spectra of all the metal complexes show 
absorption characteristic of the asymmetric (at 1600 cm" l as compared to 1670 cm" * 
in free ligand) v as COO for coordinated carboxylate group attached to the aromatic ring 
(Mishra and Jha 1980). The broad band observed in the ligand due to phenolic -OH 
reduced from 2500-3600 cm" * to 3300-3600 cm" * in complexes indicating removal of 
hydrogen bonding (Garg et al 1971). The IR spectral studies thus suggest that aromatic 
carboxylic group hydrogen is replaced and the phenolic group remains in tact. 

The analytical data agree with the general formula ML 2 (where M = bivalent metal 
ions) and ML 3 (where M = Fe(III)). TGA data show that the decomposition of the 
complexes starts at 200 (Cu), 250 (Ni), 260 (Co), 250 (Mn), 260 (Zn) and 220C (Fe) and 
complete decomposition of the organic material takes place at 550-660C leaving a 



Decomposition ^ e(r 
Compound Temp (C) (BM) 


Energies Transition 


d (CHB) PA 


220 






Cu [<5 (CHB) PA] 2 


200 


2-00 


14285 2 B lg -> ~B 2g 


Ni [d (CHB) PA-H 2 O] 2 


250 


2-71 


8770 3 A 2g (F) -> 3 r 29 (F) 








14290 3 A 2g (F)-**T lg (F) 








22222 *A 2g (F)-> 3 T lg (P) 


Co [<5 (CHB) PA] 2 


260 


3-86 


8000 , *A 2 - T, (F) 








16667 4 /l 2 -v 4 r!(P) 


Mn [<5 (CHB) PA] Z 


250 


4-42 


20000 ( M 1 - 4 T, (G) 








6/1 _v 4 P 4 d, (ft) 
/lj > ti , /^j |,UJ 


Fe [<5 (CHB) PA] 3 


220 


4-84 


90fWn 6 /4 > * A 4 F 
AA_/v/\-fv la ^ lu' o 








14285 6 /4i 9 -^ 4 r 29 








12500 6 A lg -+ 4 T lg (G) 


Zn [<5 (CHB) PA] 2 


260 


Diamagnetic 






moment to spin only value (1-73 BM) might be due to orbital contribution (Baker et al 
1966). The magnetic susceptibility data wants detailed examination because of the low 
values obtained in the case of Ni(II), Co(II), Mn(II) and Fe(III) complexes. Ni(II) 
complex shows a magnetic moment of 2-71 BM which is lower than the spin only value. 
This may be due to the octahedral distortion (Rastogi and Sharma 1974). Cobalt (II) 
complex exhibits magnetic moment of 3-86 BM which is lower than that expected for 
regular tetrahedral Co (II) complexes. The subnormal magnetic moment may be due to 
the covalent nature of the metal ligand bonds (Sengupta et al 1981). The room 
temperature magnetic moment of the Mn(II) complex is 4-42 BM which does not 
conform well with the spin only moment expected for the spin free configuration. The 
low value of magnetic moment may be due to aerial oxidation of Mn(II) -> Mn(III) 
during preparation. However, some workers explain this low magnetic moment on the 
basis of antiferromagnetic interaction between manganese(II) ions in solid state (Patel 
and Patil 1982). The effective moment of the Fe(III) complex is 4-84 BM which is quite 
lower than the spin only value of 5-92 BM for high spin d 5 system. Subnormal magnetic 
moment can be interpreted in terms of antiferromagnetic behaviour (Syamal and Kale 
1980). In most cases the electronic spectrum of Cu(II) complex possesses only a single 
broad band, making the assignment of individual electronic transition difficult (Patel 
et al 1981). A broad band at 14285cm" 1 is observed which may be assigned to 
2 B lff -> 2 B 29 transition for square planar stereochemistry (Patel et al 1981). Nickel(II) 
complex shows bands at 8770 cm" * (vj ), 14290 cm~ l (v 2 ) and 22222 cm~ 1 (v 3 ) which 
may be assigned to 3 A 2g (F)^ *T 2g (F), *A 2g (F)-^*T lg (F) and 3 A 2g (F)-* 2 T lg (P) 
transitions respectively for distorted octahedral stereochemistry (Lever 1968). The 
value of B (Racah inter-electronic repulsion parameter) was calculated using the 
method suggested by Konig (1971). The calculated values of B, /? and CFSE are found to 
be 680 cm' 1 , 0-629 and 30-08 kcal/mol respectively. The ratio v 2 /v i (1-60) supports 
distortion in octahedral stereochemistry (Arora and Misra 1982). The reflectance 



to be 895-8 cm *, 0-80 and 12-8 kcal/mole respectively. Manganese (II) complex exhibits 
a broad absorption band at 20000 cm~ x which may be assigned to the group of three 
lowest energy bands as % - % (G), 6 A 1 -* 4 r 2 (G) and % -> 4 E, % (G) for 
tetrahedral stereochemistry (Forster and Goodgame 1964). In Fe(III) complex three 
bands are observed at 20000, 14285 and 12500 cm~ ' attributable to 6 A lg -> *A lg *E g , 
6 A ig - 4 T 2g and 6 A lg -* *!T l9 (G) transitions for octahedral stereochemistry (Srivastava 
et al 1974). The Zn(II) complex is diamagnetic and may have a tetrahedral structure. 

Acknowledgements 

The authors are thankful to Prof. S R Patel, for providing laboratory facilities and 
encouragement during the course of this work. One of the authors (MRC) is grateful to 
the UGC, New Delhi for the award of research fellowship. 

References 

Arora O P and Misra S N 1982 J. Indian Chem. Soc. 59 32 

Baker E : N, Hall D and Waters T N 1966 J. Chem. Soc. 680 

Bassi P S, Gupta B R and Sharma I B 1980 Proc. Indian Acad. Sci. (Chem. Sci.) 89 125 

Forster D and Goodgame D M L J. Chem. Soc. 2490 

Garg C L, Narasimham K V and Tripathi B N 1971 J. Inorg. Nucl. Chem. 33 387 

Goyal K C and Khosla B D 1980 J. Indian Chem. Soc. 57 124 

Hariharan M and Urbach F L 1969 Inorg. Chem. 8 556 

Inoue M, Kishita M and Kubo M 1964 Inorg. Chem. 3 239 

Konig B 1971 Structure and bonding (Springer: Berlin) 9 175 

Koppikar D K. and Soundararajan S 1976 Curr. Sci. 3 41 

Lever A B P 1968 Inorganic electronic spectroscopy (Amsterdam: Elsevier) 

Mishra H C and Jha R R 1980 J. Indian Chem. Soc. 57 769 

Patel M N and Patil S H 1982 J. Macromol. Sci. Chem. All 4 675 

Patel M N and Patel M M 1981 Indian J. Chem. 20 628 

Patel M M, Patel M R, Patel M N and Patel R P 1981 Indian J. Chem. 20 623 

Rastogi D K and Sharma K C 1974 J. Inorg. Nucl. Chem. 36 2219 

Sengupta S K, Sahni S K. and Kapoor R N 1981 Indian J. Chem. A20 692 

Srivastava A K, Rana V B and Madan Mohan 1974 J. Inorg. Nucl. Chem. 36 3864 

Syamal A and Kale K S 1980 Indian J. Chem. 19 488 



Polarographic study of the complexes of cadmlum(I!) and lead (II) with 
methoxyacetate ions 

RAM PRAKASH*t, S K REHANI and RENU BALA 

Department of Chemistry, Panjab University, Chandigarh 160014, India 
t Present address: Kurukshetra University, Kurukshetra 132119, India 

MS received 1 January 1983; revised 30 April 1983 

Abstract. Reduction of the complexes of cadmium(II) and lead(II) at DME in aqueous and 
aqueous-methanol media at p. = 1-0 M(NaClO 4 ) at 15 + 0-1 and 250-1C is reversible and 
diffusion-controlled. Four complex species are formed in either case. The overall stability 
constants of 1 : 1, 1 : 2, 1 : 3 and 1 : 4 complexes have been determined. Lead (II) complexes are 
much stronger than the corresponding cadmium(II) complexes. 

Keywords. Polarography; cadmium; lead methoxyacetate. 

1. Introduction 

Recently, the plant auxins such as 3-indoleacetate, 3-indole-butyrate, and 1-naph- 
thaleneacetate have been reported to form weak complexes with trace elements 
cadmium(II), zinc(II), manganese (II) and thallium(I) (Parkash et al 1981, 1983). Since 
fungicide action is also an important phenomenon occurring in plants, it was 
considered worthwhile to study the interaction of methoxyacetate, a fungicide, with 
trace elements cadmium(II) and lead(II) at DME in aqueous and aqueous-methanol 
mixture media. 

2. Experimental 

Cadmium nitrate tetrahydrate (E Merck, AG), lead nitrate (BDH, AnalaR), sodium 
perchlorate monohydrate (E Merck, GR), perchloric acid (Reidel) and Triton-X-100 
(Rohms and Hass Co.) were used. All the other chemicals used were of guaranteed 
purity. Triple-distilled mercury and double distilled water were used. Methanol was 
purified and distilled before use (Vogel 1959). Methoxyacetic acid was prepared from 
sodium methoxide and freshly-distilled chloroacetic acid (Blatt 1946) and purified by 
distillation under reduced pressure (96-5C/l 3 mm) b.p. 203-204. Structure and purity 
of the acid was further confirmed by its PMR spectrum. 

Stock solutions of cadmium(II) and lead(II) were prepared in water and standardized 
(Welcher 1959). Sodium salt of methoxyacetic acid was prepared and pH of this stock 
solution was adjusted using ELICO pH meter model LI- 10. Solutions containing metal 
ions (4 x 10~~ 4 M) and varying concentrations of the complexing agent (0-0-0-9 M) 
were prepared in water and 10 % methanol media at ionic strength 1 M maintained with 
sodium perchlorate. 



gas presaturated with the background solution to be polarographed, was used for 
deaeration and an inert atmosphere was maintained over the solution during 
electrolysis. The electrolysis was carried out in a H-cell in conjunction with an agar-agar 
plug saturated with sodium chloride. Polarograms of 0-4 mM cadmium (II) and lead(II) 
solutions were obtained in the presence of different concentrations of methoxyacetate 
at 15 + 0-1 and 250-1C using REDELKIS polarograph type OH- 102 in aqueous and 
10 % methanol media. Triton-X-100 (0-002 % in the final solution) was used as maxima 
suppressor whenever required. An iR compensation was used while working with 
methanolic solutions. The currents were corrected for residual current. The resulting 
polarographic data as a function of the ligand concentration (C x ) which was calculated 
from the pH of the solution and pK a value of the ligand (3-55 at 15C, 3-57 at 25C 
(King I960)), are given in table 1. 

3. Results and discussion 

Both cadmium(II) and lead(II) give single well-defined reversible polarographic 
reduction wave. Current is directly proportional to (h eK ) 1/2 showing that the reduction 
is diffusion-controlled. Plot of de vs log i/i d i is a straight line with a slope of 32 
1 mV. Cathodic reduction of cadmium(II) and lead(II) in methoxyacetate is, 
therefore, reversible and involves two electron transfer in either case. That E II2 is 

Table 1. Polarographic data of the interaction of cadmium(II) and lead(II) with meth- 
oxyacetate ions. 



Cadmium (Il)-methoxyacetate system in Lead (II) methoxyacetate system 

aqueous medium at 10% methanol medium at in aqueous medium at 



25 ore 



1501C 



25(MC 25 + (MC 



15 orc 



M 




El/2 

(-V) 


id 

(M) 


1/2 
(-V) 


*i 

(M) 


1/2 
(-V) 


id 
0*A) 


Ej/2 

(-V) 


id 
(M) 


-El/2 

(-V) 


id 
(M) 


0-0000 





0-5800 


1-872 


0-5865 


1-584 


0-5695 


1-733 


03755 


2-088 


0-3810 


1-992 


0-0200 


1-699 


0-5810 


1-814 


05885 


1-464 


0-5725 


1-570 


0-3855 


2-047 


03905 


1-934 


0-0500 


1-301 


0-5825 


1-800 


05920 


1-459 


0-5770 


1-546 


0-3965 


1-824 


0-3985 


1-934 


00999 


1-000 


0-5860 


1-680 


0-5970 


1-428 


0-5838 


1-464 


0-4025 


1-872 


0-4100 


1-728 


0-1499 


0-824 


0-5895 


1-584 


0-6010 


1-392 


0-5890 


1-440 


0-4145 


1-853 


0-4160 


1-680 


01999 


0-699 


05945 


1-661 


06045 


1-356 


0-5943 


1-450 


0-4210 


1-862 


0-4215 


1-618 


0-2499 


0-602 


05980 


1-632 


06080 


1-363 


0-5985 


1-416 


0-4250 


1-824 


0-4250 


1-363 


0-2998 


0-523 


06015 


1-560 


0-6108 


1-344 


0-6025 


1-392 


0-4290 


1-781 


0-4280 


1-176 


0-3498 


0-456 


0-6050 


1-555 


0-6130 


1-272 


0-6060 


1-356 


0-4330 


1-728 


0-4345 


1-416 


0-3998 


0-398 


0-6080 


1-546 


0-6160 


1-272 


0-6100 


1-392 


0-4360 


1-651 


0-4380 


1-354 


0-4497 


0-347 


0-6115 


1-536 


0-6183 


1-262 


0-6135 


1-380 


0-4400 


1-651 


0-4405 


1-248 


0-4997 


0-301 


0-6148 


1-519 


06205 


1-272 


0-6155 


1-320 


0-4425 


1-560 


0-4430 


1-145 


0-5996 


0-222 


06205 


1-488 


06250 


1-224 


0-6213 


1-308 


0-4485 


1-560 


04485 


1-056 


0-6996 


0-155 


06255 


1-440 


0-6298 


1-248 


0-6265 


1-248 


0-4540 


1-512 


0-4540 


1-114 


0-7995 


0-097 


06300 


1-368 


06325 


1-152 


0-6310 


1-200 


0-4588 


1-440 


0-4595 


1-145 


0-8995 


0-046 


0-6335 


1-296 


0-6355 


1-128 








04600 


1-157 


0-4630 


0-984 



Polarographic study ofCd(II) and Pb(II) complexes 



71 



independent of the effective height of mercury column (h efr ) is in keeping with the 
reversible character of the electrode process. There is an appreciable cathodic shift in 
the 1/2 with increase in ligand concentration in either case. The diffusion current also 
decreases with increasing concentration of the complexing agent due to increase in the 
size of Cd(II) and Pb(II) ions on complex formation. 

F (X) functions are calculated at different concentrations of the ligand (DeFord 
and Hume 1951) and then solved for the stability constant by a graphical procedure, 
depicted for cadmium as a typical example, in figure 1, which confirms the formation of 
four consecutive complexes, 1:1, 1:2, 1:3 and 1 : 4 in equilibrium with each other in 
either case, having stabilities for the cadmium-methoxyacetate complexes ^ =5, 13, 
17; 2 = 25-5, 29, 67, 3 = 37, 20-5, 108-5 and 4 = 63-48, 40-25, 149-4 at 25 0-1, 15 
0-1C (both in aqueous medium) and 25 0-1 C (in 10 % methanol) respectively; for 
lead-methoxyacetate complexes in aqueous medium ^ = 70, 80; jS 2 = 450, 200; /? 3 
= 210, 900 and 4 = 1120-96, 861-79 at 25 0-1 and 15 0-1C respectively. Lead(II) 
complexes with methoxyacetate could not be studied in methanol medium due to 
precipitation of the complexes formed. The values of overall stability constants 
calculated by Mihailov's method (1974) are in close agreement with those calculated by 
DeFord and Hume's method. 

Percentage distribution of cadmium(II) and lead(II) present in various forms in 
equilibrium as a function of logarithm of ligand concentration has also been calculated 
and a typical figure (figure 2) for distribution diagrams of this type is given. 





1-7 1.5 



1.0 0.5 

-log C x (methoxyacetate) 



Figure 2. Percentage distribution of cadmium in various forms as a function of log C x ir 
aqueous medium at 25O1C. 

It is evident from the results that the complexes formed are more stable at lowei 
temperatures or in methanol medium. However, the highest complex becomes less 
stable in both cases on lowering the temperature. The lead (Il)-methoxy acetate 
complexes are, in general, more stable than the corresponding cadmium(II) complexes 

Acknowledgement 

One of the authors (RB) is thankful to UGC, New Delhi for the award of a fellowship 



References 

Blatt A H 1946 Org. Synth. Coll. 2 260 

DeFord D D and Hume D N 1951 J. Am. Chem. Soc. 73 5321 

King E J 1960 J. Am. Chem. Soc. 82 3575 

Mihailov M H 1974 J. Inorg. Nucl. Chem. 36 107 

Parkash R, Singh B and Bala R 1981 Trans. Soc. Adv. Electrochem. Sci. Technol. 16 141 

Parkash R, Singh B and Bala R 1983 J. Indian Chem. Soc. 60 92 

Vogel A I 1959 A textbook of practical organic chemistry (London: Longman Green) 

Welcher F J 1959 Analytical uses of ethylenediaminetetraacetic acid (New York: Van Nostrand) 



Abstract: 

All papers must have an abstract of not more than 
200 words of the significant results reported in the 
paper. 

Keywords: 

Between 3 and 6 keywords must be provided for 
indexing and information retrieval. 

The text: 

The paper must be divided into sections starting 
preferably with 'Introduction' and ending with 
'Conclusions'. The main sections should be num- 
bered 1, 2, etc, sub-sections 1.1, 1.2, etc., and further 
sub-sections (if necessary) 1.1 a, l.lb, etc. 

Tables: 

All tables must be numbered consecutively in arabic 
numerals in the order of appearance in the text. The 
tables should be self-contained and have a descript- 
ive title. 

Figures: 

All figures including photographs should be num- 
bered consecutively in arabic numerals in the order 
of appearance in the text. The figure captioris must 
be typed on a separate sheet. 

Line drawings must be in Indian ink on good 
quality tracing paper or on white card of a size 
smaller than that of text paper. Structural formulae 
of compounds must be drawn in Indian ink on 
tracing paper. Lines must be drawn sufficiently thick 
for reduction to a half or third of the original size 
(0.6mm for axes and 0.3 mrn for curves are 
suggested). 

List of symbols: 

For the convenience of the printers, authors should 
attach to the manuscript the complete list of sym- 
bols identified typographically. 

Mathematical material: 

Equations must be clearly written, each on its own 
line, well away from the .text. All equations must be 
numbered consecutively in arabic numerals with the 
number in parentheses near the right hand margin. 
Indices or superscripts must be indicated in pencil 
with a ^ sign and subscripts with a /^ sign, 'oh' and 



colour may be used to denote (lower case respect- 
ively capital) characters blue for Greek, red for 
German (gothic), green for script and brown for 
bold face. Authors must indicate wherever special 
characters (Greek, German, script, scalar, vector, 
tensor, matrix, etc.) are required. All other letters 
will be set in italic types. Vectors must be underlined 
by a wavy line and tensors by two wavy lines. The SI 
system of units and symbols is recommended. 

References: 

References should be cited in the text by author and 
year. However, in the case of mathematical sciences, 
the references in text should be by serial number 
only (e.g. [1]). If there are more than two authors, 
reference should be to the first author followed by et 
al in the text. References at the end of the paper 
should be listed alphabetically by authors' names, 
followed by initials, year of publication, name of the 
journal (abbreviated according to the World List of 
Scientific Periodicals, Butterworths, London), 
volume number, and number of the first page. 
References to books should include: name(s) of 
author(s), initials, year of publication, title of the 
book, edition if not the first, initials and name(s) of 
editor(s) if any, preceded by ed(s), place of publi- 
cation, publisher and chapter or pages referred to. 
References to thesis must include the year, the title 
of the thesis, the degree for which submitted and the 
University. 

Footnotes: 

Footnotes to the text should be avoided if possible 
but when necessary should be numbered consecut- 
ively, and typed on a separate sheet. 

Appendices: 

All appendices should be numbered consecutively in 
arabic numerals. 

Proofs: 

Authors are requested to prepare the manuscript 
carefully before submitting it for publication to 
minimise corrections and alterations in the proof 
stage which increase publication costs. The proofs 
sent to the authors together with the reprint 
order form must be returned to the editorial office 
within two days of their receipt. 

Reprints: 

50 reprints of each article will be supplied free of 



Chemical Sciences 

CONTENTS 

Physical and Theoretical 

Absorption and emission spectra of isomeric tolunitriles 

, A Maiti, S K Sarkar and G S Kastha 

Spectrophotometric determination of basicities of substituted acetylbiphenyls 
and biphenyl carboxyiic acids P Ananthakrishna Nadar and N Kannan 

Effect of substituents on the oxidation of some alkyl-aryl sulphoxides by 
chloramine-T K Ganapathy and P Jayagandhi 

Spectroscopic studies of the electron donor-acceptor interactions of aromatic 

hydrocarbons with tetrachlorophthalic anhydride 

PC Dwivedi and Avanija Gupta 

A zero differential overlap study of chemical binding in ammonia-borane and 
carbonyi-borane Dulal C Ghosh 

Organic 

Reaction of the carbonate radical with substituted anilines . 

T P Elango, V Ramakrishnan, S Vancheesan and J C Kuriacose 

Boron complexes of sulphur containing Schiff bases derived by the condensation 

of S-methyl or S-benzyl dithiocarbazate with jS-diketones 

, . P K Singh and J P Tandoh 

Reduction mechanism of l,4-diamino-2,3-anthraquinone disulphonic acid . . 
F Capitan, E Alvarez-Manzaneda and J L Vilchez 

Inorganic and Analytical 

Spectral, magnetic and thermal studies of transition metal complexes of 

<5(3-carboxy, 4-hydroxy benzoyl) pentanoic acid 

. M N Patel and M R Chaudhari 

Polarographic study of the complexes of cadmium(II) and lead(II) with 
methoxyacetate ions Ram Prakash, S K Rehani and Renu Bala 



Indexed in CURRENT CONTENTS 



Edited and published by C N R Rao for the Indian Academy of Sciences, Bangalore 560080. Typeset by 
Macmillan India Ltd., Bangalore 560025 and printed at the Macmillan India Press, Madras 600041. 



Kinetic studies on the homogeneous hydrogenation of fumaric and 
maleic acids catalysed by bis(dimethy!glyoximato)cobalt(II) 

' (Mrs) S VASANTHKUMAR, S VANCHEESAN, J RAJARAM AND 
J C KURIACOSE* 

Department of Chemistry, Indian Institute of Technology, Madras 600 036, India. 
MS received 20 May 1983; revised 3 September 1983 

Abstract. The influence of varying concentrations of Co(DMOH) 2 , NaOH and axial base on 
the rate of hydrogenation of fumaric and maleic acids has been studied in detail. Intramolecular 
hydrogen bonding in the monoanion of maleic acid and the trans orientation of carboxylic acid 
groups in fumaric acid are important factors which account for the difference in the rate of 
hydrogenation of these substrates. Mono-, di- and trialkyl amines as axial bases modify the 
activity of the catalyst, dialkylamines conferring the maximum activity and trialkylamines the 
least. Back-strain on nitrogen atom and solvation energy of the amines are responsible for their 
different behaviours. A rate law has been proposed and verified. 

Keywords. Cobaloxime; hydrogenation; axial base; fumaric acid; maleic acid. 



1. Introduction 

Glyoxime complexes of cobalt(II) accept bases in the axial positions to form 
cobaloximes (Schrauzer 1968;Schrauzerand Windgassen 1970; Schrauzerand Holland 
1971; Belluco 1970). Because of the similarity to Vitamin B J2 in their reactions they 
serve as excellent models for Vitamin B 12 (Schrauzer 1976). During the past decade 
there has been considerable growth in the studies related to the bioinorganic aspects of 
Co(II). These studies involve synthesis and reactions of complexes of Co(II) with 
dimethyl and other glyoximes, Schiff bases and other macrocyclic ligands. The green 
Co(I) species formed by the reduction of cobaloximes have strong reducing properties 
and nucleophilic character (Dodd and Johnson 1973; Witman and Weber 1977). As a 
part of our studies on the activation of molecular hydrogen by complexes (Vancheesan 
et al 1978; Rajagopal et al 1979; Pillai et al 1980, 1982; Thangaraj et al 1980) we have 
investigated the hydrogenation of fumaric and maleic acids by cobaloxime. We report 
here the results of our studies on the role of (i) orientation of functional groups as in 
fumaric and maleic acids, (ii) the influence of the axial base in the complex and (iii) the 
concentration of NaOH on the rate of hydrogenation. 

2. Experimental 

Cobalt chloride hexahydrate and dimethylglyoxime (BDH,AR) were used for the 
preparation of cobaloxime (III). The catalyst solutions were prepared in situ in aqueous 



shaking with a saturated solution of ferrous ammonium sulphate (BDH,AR). Al] 
cobaloxime solutions were saturated with hydrogen at the reaction temperature 
to the addition of substrates by vigorous stirring till no further hydrogen uptak 
observed. The hydrogen absorption was followed at regular intervals by means oi 
burette with provision to maintain a constant pressure of hydrogen. Kinetic si 
were carried out by changing any one of the following parameters at a time ke 
others constant: (i) catalyst concentration, (ii) substrate concentration, (iii) axial 
(iv) concentration of sodium hydroxide and (v) temperature. Reaction mixtures 
withdrawn periodically for product analysis. After separation from the catalys 
other reagents, solid products were analysed by TLC technique using alumina. L 
samples were analysed by GLC using Varian 1800 model with hydrogen as carrie 

3. Results and discussion 

3.1. Effect of axial base 

Co (oMGH) 2 forms 1 : 1 and 1 : 2 adducts with a variety of Lewis bases. These adduc 
soluble in a number of organic solvents and water. 1 : 2 adducts dissociate in soluti 

Co(DMGH) 2 2B - Co(DMGH) 2 B + B 

Under hydrogenation conditions hydrido cobaloximes, HCo(DMG) 2 B (DMG = n 
anion of dimethylglyoxime) which are Bronsted acids corresponding to the 
nucleophiles, are formed. This hydrido species is the catalytic intermediate as in 
well-known examples of Ru(II), Rh(I), etc. (Hallman et al 1968; Osborn et al 1966 
stability of the hydridocobaloximes is very much influenced by the axial 
(Schrauzer and Holland 1971). Predominantly a donor nitrogen bases form uns 
hydrido species that cannot be isolated whereas Ti-acceptors like triaryl and tri 
phosphines as axial ligands form stable hydrides (Schrauzer and Holland 1971). '. 
hydride intermediate is too stable, migration of hydride to coordinated subs 
becomes difficult, rendering the complex catalytically inactive. On this account pyi 
and other N-donor ligands like various substituted amines are considered better 
bases than phosphines for catalytically active complexes. Among several niti 
containing compounds (table 1) tried, pyridine was found to be the best. Ther 
several reports of pyridine being a good axial base in cobaloxime catalysis (Simand 
1980; Zahonyi et al 1972, 1975). 

The catalytic activity of the cobaloxime is attributed to the presence of unpai: 
electrons in Co(II) which is extensively delocalised along the z axis as indicated b 
studies (Simandi et al 1972, 1975). The result of such delocalisation is an increase i 
free radical nature of Co(II), facilitating the interaction with hydrogen throu 
homolytic cleavage of molecular hydrogen. The efficiency of the axial base rm 
related to the availability of the lone pair. Thus among the mono, di and trialkyl ac 
the following order of reactivity is expected: 

trialkylamine > dialkylamine > monoalkylamine 

r thf initial l-Qto-c 



Base 


H 2 absorbed 
"(ml/hr) 


% conversion 
to product 


n-butylamine 


8-1 


33 


di-n-butylamine 


12-0 


46 


tri-n-butylamine 


nil 


nil 


Triethyl amine 


nil 


nil 


Trimethyl amine 


nil 


nil 


Pyridine 


25 


98 



"^Concentrations of catalyst, substrate, NaOH, axial base and tempera- 
ture are maintained constant. 

reactivity. This deviation from expected reactivity could be attributed to increased 
back-strain on the N-atom in trisubstituted amines which increases the 
p-character on the lone pair. However, in solution the basicity is increased with the 
extent of solvation and the solvation energies of the conjugate ammonium ions are in 
the order: 

RNH 3 + > R 2 NH 2 + > R 3 NH + 

Electronic and solvation effects which oppose each other decide the basicity. As a result 
dilakylamine has maximum basicity. The difference in basicity is reflected in the activity 
of the complex for hydrogenation which is in the decreasing order: 

disubstituted > monosubstituted > trisubstituted 
With trialkylamines as axial base there was no absorption of hydrogen. 

3.2 Effect of alkali concentration 

In a strongly alkaline medium (pH > 9), cobaloxime exists predominantly as Co (I) and 
Vitamin B 12 as B 12s (equations (1) and (2)) (Schrauzer et al 1965; Schrauzer and 
Windgassen 1966; Hill et al 1969). 



HCo(DMGH) 2 B + OH- ^[Co'(DMGH) 2 B]- + H 2 O (1) 

-fOH" 

2 Vitamin B 12r - - x Vitamin B t 2s + Vitamin BI 2fl (2) 

-OH~ 

(In (2) subscripts s, r and a indicate the oxidation states of cobalt.) In Co 1 the highest 
occupied orbital is probably a weakly antibonding d z 2 orbital which forms the centre of 
high polarizability and charge density. Since the equilibrium will be shifted to the right 
in strongly alkaline medium (equations (1) and (2)) the presence of the acid form 
HCo (III) is unlikely (equation (1)). In the presence of H 2 at 1 atmosphere pressure and 
pH > 9, Co (I) is a very stable form (Takeuchi and Ohgo 1974). However at pH between 
7 and 9, HCo (1II) is the predominant species and the hydrogenation occurs essentially 
through proton transfer, the mode of addition of substrate being cis. However if Co (I) 
species reacts with the substrate forming an unstable a complex, which is subsequently 
converted to a stable n complex, its further hvdroeenation can be effected onlv under 



Temperature 31 C 

[Catalyst] [Maleic acid] = [Fumaric acid] = 2.0 mM 

1.0 mM O 

L 2.0 mM D B 




[NaOH] mM 
Figure 1 . Effect of [NaOH] on the rate of hydrogenation of fumaric acid and maleic acid. 



alkaline conditions the active form of cobaloxime is HCo(DMGH) 2 B. Since the 
substrates used are carboxylic acids, some amount of alkali is used up for neutralizing 
the acid. The effective [OH~] is the excess remaining after neutralisation. On 
increasing [OH~] the rate of hydrogenation passes through a maximum 
(Shanthalakshmy et al 1980). This is in agreement with Schrauzer's scheme where 
cobaloxime(II) is converted into cobaloxime(III) and cobaloxime(I) (Schrauzer et al 
1965, 1966, 1968). 

i r\Lj - 




This reaction is complete in the presence of excess alkali which explains the steep fall in 
the rate as the concentration of alkali is increased. The order with respect to catalyst and 
the substrate is one each and that with respect to [OH~] is fractional. 

3.3 Nature of the substrate 

The rate of hydrogenation of maleic acid was found to be higher than that of fumaric 
acid. On adding OH", maleic acid forms the monoanion which enters into hydrogen 
bonding with the other carboxyl group which is in close proximity to the carboxylate 
anion. Hence under the experimental conditions further neutralisation of maleic acid to 



iimuwiiig. \i) me jtinjiiDctiiiuii ui mcuci^ ai,iu wmi an avaiiauic n. 

formation of hydridocobaloxime(III) which is the active intermediate in the reaction. 
This is analogous to transfer of hydrogen from a donor solvent to an acceptor via 
hydride intermediate (Pillai et al 1980, 1982). The maleate monoanion probably 
functions as a hydride donor too. This is an additional factor which enhances the rate. 
In fumaric acid, dianion formation does not leave any scope for the substrate to 
function as a hydrogen donor. Besides, the trans orientation of carboxylate groups is a 
less favourable geometry for the activation of substrate. This accounts for the difference 
in the rates of hydrogenation of the two substrates. 

3.4 Rate law 

The following reaction sequence may be considered. 

(3) 
HCo< HI > . (4) 

fc-2 

HCo< in 'B + OH" S Cow B- + H 2 (5) 

USH + OH- S=US-+H 2 O (6) 

HCo< ni >B + US ~ -4- HCo< UI >B US ~ + H 2 - Products (7) 

(8) 






Assuming, * = K' 4 , (10) becomes, 



[US'] = Ki[USH][OR-] (11) 



tK'i [Co(">] [B] [H,] 1 / 2 [USH] [OH"] 



The observed orders with respect to catalyst and substrate are in agreement with the 



ueiiomuicuui 111 ^i*+j wa.ii uc iicgict-icu, leaving umy /t_ 2 . ML vciy mgu |_v_/n j, icox/uuu 

(5) is predominant as discussed earlier. This can be considered to occur via formation of 
Co ll B as an intermediate (equation (3)). With increase in [OH~], X 3 [OH~] and 
/c 5 K"4[USH][OH~] increase, since in the denominator of (14), [OH~] becomes 
considerable compared to k_ 2 , causing a steep fall in the rate on increasing [OH~]. At 
low [OH~], -K 3 [OH~] may be small compared to the rest of the terms in the 
denominator of (14). Neglecting this and rearranging we get 

[Co 11 ] [B] [USH] _ fc_ 2 [USH] 



R;. is i \. is [~TJ "1 1 
Kj.tY4K 2 A.i |_Jn.2J 

A plot of the left side of the above equation against 1/[OH~] is almost parallel to the 
y axis. Hence the term X 3 [OH~] is included and (14) can be rearranged to give (15) 

[USH][Co n ][B]_ fc_2 K 3 

+ k K j- H -11/2 O 5 ) 

If fc 2 [H 2 ] 1/2 is represented as k' 2 ,k s k' 2 KiK' 4 as K and [Co 11 ] [B] [USH]//? as A, (15) 
can be rewritten as 

A _ k. 2 K 3 t [USH] 



~ JC[OH-] K k' 2 K, 
The variation in A as [OH~] is changed at constant [USH] is given by 

dA _ -fc. 2 
d[OH'] ~K[OH~] 2 

dA k 



From the intercept of a plot of left side of (17) against In [OH"], k, 2 /K has been 
evaluated (table 2, figures 2 and 3). In order to evaluate k' 2 K t , (16) is rearranged after 
substituting for A and written as (18) 

[Co"][B] 1 



R 



Table 2. Evaluation of some kinetic parameters. 











Energy of 
activation 




k-i/K 


k' 2 Kl 


KS/K* 


(kcal/mol) 


Maleic acid 


0-0264 


1-19 


52-5 


17-6 


Fumaric acid 


0-1114 


1-00 


100 


21-3 




-2 



Figure 2. Plot ofln(d/4/d[OH]~)t>sln[OH]~ for the evaluation of -k - 2 /K A for fumaric 
acid. Verification of eq. (17). 



'S 
G, 







0-5 



Temperature 31 C 

[Catalyst] [F^aleic acid] 
mM 
1.0 



2.0 
1.5 
1.5 
1.0 



mM 
2.0 
2.0 
1.0 



1.0 

TniJl" 



1.5 



2.0 



u ^ vasanmKwnar et ai 

Differentiating (18) with respect to [OH""], we get 

f[CO"][B] 1 |d[USH] f[Co<")][B][USH]| 

\ R fc'aJ^JdtOH-] I R 2 j 

dR = -fe-a 
d[OH'] ~K[OH-] 2 

Rearranging (19) we get 

d[USH] [[Co<">][B][USH] dfl /c_ 2 

d[OH~] * 2 j R 2 d[OH-] K[OH~] 2 

[Co' 11 '] [B] d[USH] | 

K d[OH~]J * 2 ^ 

where Q stands for the quantity in flower bracket { } of (20). Q can be calculated 
the known value ofk- 2 /K and other terms. The linear plot of left side of (20) aga: 
passes through the origin and from the slope of this plot, the Kik' 2 value has 
evaluated (table 2). The K^k' 2 and k, 2 /K values are substituted in (18) and the 
of K 3 /K is obtained from the intercept of a plot of left side of (18) against 1/[C 
The K 3 /K values are given in table 2. The large differences in the k- 2 /K values ai 
for the two acids might be due to the differences in JC 4 (ionization constants of th 
acids) and k s . The lower rate of hydrogenation of fumaric acid compared to malei 
suggests that k s might be small for fumaric acid. The k- 2 /K and K 3 /K values art 
larger for fumaric acid. 



References 

Belluco U 1970 Inorg. Chim. Act. Rev. 4 41 
Brauman J I and Blair L K 1968 J. Am. Chem. Soc. 90 6561 
Brown H C 1945 J. Am. Chem. Soc. 67 374, 378 
Dodd D and Johnson M D 1973 J. Organometal. Chem. 52 1 
Hallman P S, McGarvey B R and Wilkinson G 1968 ./. Chem. Soc. A 3143 
Hill HAD, Pratt J M and Williams R J P 1969 Chem. Br. 5 156 
' Munson MSB 1965 J. Am. Chem. Soc. 87 2332 

Muthukumaru Pillai S, Vancheesan S, Rajaram J and Kuriacose J C 1980 Indian J. Chem. A19 1C 
Muthukumaru Pillai S, Vancheesan S, Rajaram J and Kuriacose J C 1982 .7. Mol. Catal. 16 349 
Ohgo Y, Takeuchi S and Yoshimura J 1971 Bull. Chem. Soc. Jpn. 44 283 
Osborn J A, Jardine F H, Young J F and Wilkinson G 1966 J. Chem. Soc. A 1711 
Rajagopal S, Vancheesan S, Rajaram J and Kuriacose J C 1979 Indian J. Chem. B18 293 
Ricroch M N and Gaudemer A 1974 J., Organometal. Chem. 67 119 
Schrauzer G N and Holland R J 1971 J. Am. Chem. Soc. 93 1503 
Schrauzer G N 1968 Ace. Chem. Res. 1 67 

Schrauzer G N and Windgassen R J 1970 J. Am. Chem. Soc. 92 1551, 2997, 7078 
Schrauzer G N 1976 Angew. Chem. Int. Ed. 15 417 
Schrauzer G N 1977 Angew. Chem. Int. Ed. 16 233 



Simandi L I, Zahonyi E B, Szeverenyi and Nemeth S 1980 J. Chetn. Soc. (Dalton) 276 

Takeuchi S and Ohgo Y 1974 Bull. Chem. Soc. Japan 47 463 

Thangaraj T, Vancheesan S, Rajaram J and Kuriacose J C 1980 Indian J. Chem. A19 404 

Vancheesan S, Prabha Sethi S, Rajaram J and Kuriacose J C 1978 Indian J. Chem. A 16 399 

Witman M W and Weber J H 1977 Inorg. Chim. Acta. 23 263 

Zahonyi E B and Simandi L I 1972 J. Coord. Chem. 2 53 

Zahonyi E B and Simandi L I 1975 J. Chem. Soc. (Dalton) 1729