thermal stabilisation of polyvinylchloride: effect of rubber seed oil
TRANSCRIPT
Journal of Scientifi c & Industri al Research
Vol. 59,July 2000, pp 563-568
Thermal Stabilisation of Polyvinylchloride: Effect of Rubber Seed Oil Derivatives on the Thermal Degradation of Polyvinylchloride
FE Okieimen
Uni versi ty of Benin, Departmen t of Chemistry, Benin City, igeria
Received: 2 1 October 1999; accepted: 27 April 2000
Thermal degradation of polyvinylch loride is studied under non-oxidati ve and oxidat ive condi ti ons in the presence of cpoxid iscd ru bber seed oil and meta l soaps of the oil. Changes in in trin sic viscosity and in the leve ls of unsaturation in the degraded polymer samples arc used to assess the effect of the rubber seed oi l derivati ves on the thermal degradation or polyvinylchloride. I t is found that the metal soaps or cpoxidised rubber seed oil marked ly reduce the level of unsaturat ion in the degraded polymer sample and the ex telll of polymer chain scission which is accompanied by dehydrochlorination. The stab ilising ability of the rubber seed oil derivatives is compared with that or Jatropha seed oil.
Introduction
Polyvi nylchloride (PVC), is an important commercial polymer. A major disadvantage in the manufacture and use of PVC is its low thermal stability. At elevated temperatures we ll below its decomposition temperature, PVC loses HCI and discolourises, lead ing to deterioration in the properties of the polymer. Several workers 1 · 1 ~ have reported on the degradation and stabilisat ion of PVC. T hermal degradation of PVC is generally considered to be initiated at unstable si tes, particularly tertiary and allylic chlorides, within the polymer. Although structural irregularities with in PVC considerably increase the initial rate of e limination of HCI
1) it has
been argued that on account of the low concentration in normal PVC of these structural irregul arities that initiation of thermal degradation also takes place at
H regular monomer sequence · .
The poor thermal stability of PVC requires the use of heat stabilisers in the processing of the polymer. Heat stabilis ing additives have been suggested to prevent or retard dehydrochlorination in a several ways. Their proposed mechanism of action
16 includes
HCI scavenging and the [eplacement of labile chlorine atoms within PVC with stabiliser moieties that are more stable to heat treatment. The latter mechani sm was first proposed by Frye and Horst
17 Add itives that
have found pract ical application as thermal stabili ser
for PVC include metal soaps of carboxy lic acids , organometallic compounds and inhibitors of radical chain reactions. More recently, interest has been shown in the use of epox ides 18
"19 as co-stabili ser in PVC
formu lati ons. This study is a part of an on-going study aimed at the development of value-added products from local resources and evaluates the application of epox idi sed rubber seed oil and its metal soaps as thermal stabili ser for polyvinylchloride.
Materials and Methods
Polyvi nylch loride (BDH) (Mn 1.0 x I 0); particle size I 00 per cent pass BS 60 mesh, 74 per cent pass BS 200 mesh) was purified by solut ion in THF/acetone mixture and precipitation with constant stirring in excess of methanol. The precipitated polymer was filtered off after 24h, washed with methanol and airdried. Rubber seed oil (RSO) was obtained from the Rubber Research Institute of Nigeria, Benin. The oil was mechanically extracted from seeds collected within a period of ly. The ,Physico-chemical and fatty acid composition of RSO (ref. 20) is shown in Table I.
Epoxidation of Rubber Seed Oil
Epoxidation of RSO was carried out at 29°C using peracetic acid prepared in situ by reacting hydrogen perox ide (30 per cent v/v) with glacial acetic acid. The level of epox idation was determined using the method described by Durbetaki21
. The epoxidised
564 J SCI .IND RES VOL 59 JULY 2000
Tab le I -Physico-c hemi cal properties and ratty acid
composition or rubber seed o il
Property/Component Value
Specific gravity (at 3011C) 0.926
Free ratty acid per cent (as oleic ac id)
Acid value (mg KOH/g)
Peroxide value (meq/kg)
Iodi ne value (g li iOOg)
Fatty ac id composition (per cent)
Myristic acid Cl4:0
Palmitic ac id Cl6:0
Stearic ac id C 18:0
Oleic ac id Cl8:0
Linoleic acid C 18:2
Linolenic ac id Cl8 :3
11 .29
23.00
0.40
155.56
2.2
7 .6
10.7
20.0
36.0
23.5
rubber seed oil, (ERSO), was stored at rc until required. Detailed kinetic stud ies on the epoxidation of RSO and subsequent cleavage of the oxirane ring by acetic acid have been reported by Aigbodion eta/. 22
.
Preparation of Metal Soaps o.lRubber Seed Oil
Metal soaps were prepared from RSO and ERSO by metathesis in alcohol solution23
'24
. The sod ium soaps of the oils were first prepared by dissolving the oi l sample (9.2 g) in 50 ml of hot ethanol followed by treatment with 20 ml of 20 wt per cent sodium hydroxide solution. To thi s mixture, I 00 ml of 30 wt
per cent so lution of the metal salt were slowly added with continuous stilTing. The precipitated meta l soap was washed with hot water and air-dried.
Degradation
Thermal degradation was carried out using PVC powder in the presence of I 0 wt per cent RSO and ERSO and 3 wt per cent of the metal soaps of the oils. The polymer sample ( 1.5 g) was mixed thoroughly with the appropriate amount of the add iti ve and transferred to a degradation tube. The tube was connected to a source of nitrogen maintained at a flow rate of 60 ml/min . The degradation tube was immersed in a thermostat o il bath controlled at 190 ± I 0C. Degradation was allowed to proceed for 30 and 60 min at the end of which the degradation tube was cooled and the contents dissolved in cyclohexanone and poured into excess of methanol. The precipitated PVC was filtered off, washed several times with methanol , dried in a vacuum at room temperature, and stored in 1l1e dar:·.
Thermoxidative degradation studies were carried out in air as described for non-oxidative degradation.
Viscosity Measurements
Viscosity measurements of undegraded and degraded PVC samples were carried out 111
cyclohexanone solution at 30°C. the intrinsic viscosities and Huggins' interaction constant of the polymer samples were dete rmined usmg the relationship:
The ratios of the intrinsic viscosity of PVC samples degraded in the presence of the rubber seed oil de rivatives to the viscosity of the undegraded polymer
sample, [11]1[11lo, were used to assess the effect of the additives on the degradation of PVC.
Levels of Unsaturation in Polymer Samples
The levels of unsaturation in the undegradcd and degraded PVC samples were estimated from their iodine values dete rmined by the Wij's method25
.
Results and Discussion
Non-oxidative Degradation
The va lues of intrinsic viscosity and Huggins' interaction constant for the degraded PVC samples are shown in Table 2. It has been reported26 that the intrinsic viscosity of PVC undergoing the rmal degradation decreases initially to a minimum and then increases with further increase in the number of conjugated double bonds in the polymer. T herefore , if the ex tent of dehydrochlorination is kept low, changes in the value of intrinsic viscosity can be taken as estimates of the ex tent of degradation in which case, the values of re lative intrinsic viscosity [11]1[11] 0 , where [11]o is the intrinsic viscosity of undegraded PVC, can be used to assess the effect of rubber seed oi l derivatives on the degradation process. The reactions
that may accompany thermal dehydrochlorination of PVC include chain sc ission and cross-linking; the latter leadi ng to [11]1[11]o values > I and the former to values < I . Thus, closer the values of relative intrinsic visco~ity to unity the greater the stabi lising ab ility of the additive. The results in Table 2 show that the values of [11]1[11]o are< I, suggesting that chain sc iss ion is the predominant reaction accompanying therma l
OKIEIMEN: THERMAL STABILISATION OF POLYVINYLCHLORIDE 565
Table 2- Viscosity parameters for PVC degraded at 190°C in the presence of rubl:1er seed oil derivat ives
Additive Period of Intrinsic viscosity Huggins' Relative intrinsic
None
RSO ( I 0 wt per cent)
ERSO ( I 0 wt per cent)
48 per cent epoxidation
I 0.6 per cent epoxidation
Metal soaps of RSO (3 wt per cent)
Barium soap
Cadmium soap
Lead soap
Metal soaps of ERSO (3 wt per cent)
Barium soap
Cadmium soap
Lead soap
degradation
(min)
30
60
30
60
30
60
30
60
30
60
30
60
30
60
30
60
30
60
30
60
[T\J.. = lnlrinsic viscosity of undegraded PVC= 1.10
dehydrochlorination of the PVC samples . Table 2 shows that the values of relative intrinsic viscosity are highest for PVC degraded in the presence of metal soaps of epoxidised rubber seed oil and lowest for the sample degraded in the absence of the additives which indicates a stabilising effect. Results reveal (Table 2) that the stabilising ability of the metal soaps of epoxidised rubber seed oil is higher than that for the soaps of the unepoxidised oil. The metal soaps of the epoxidised rubber seed oil are thought to exhibit combined effects of the epoxy group (HCI scavenger) and of the metal soap (subsritution of labile chlorine atoms within PVC additive moieties) in stabilising PVC against thermal degradation .
The levels of unsaturation in PVC samples degraded at 180°C for 30 min in the presence of the rubber seed oil derivatives are shown in Table 3. The iodine values for PVC samples degraded in the
[T\] constant viscosity
(d l I g) K [T\] /[11].,
0.45 35.70 0.41
0.'36 39.54 0.33
0.67 18 .80 0 .61
0.40 42.25 0.36
0.72 14.42 0.65
0.41 38.65 0.37
0.81 12.35 0.74
0.52 25 .07 0.47
0.80 12.59 0 .73
0.64 16.88 0.58
0.69 17.85 0.63
0.57 21.88 0.52
1.00 6.88 0.91
0.68 14.78 0.62
0.90 8. 11 0.82
0.64 16.88 0.58
0.90 8.64 0.82
0.59 21.63 0.54
0.99 7.02 0.90
0.72 13.27 0.65
presence of rubber seed oil derivatives varied from 58.17 for PVC stabilised with epoxidised rubber seed oil (with 4.6 per cent epoxide content) to 29.08 for PVC stabilised with lead soap of rubber seed oil. These results represent a decrease of between 5.83 per cent and 52.91 per cent in iodine value relative to the value obtained for unstabilised PVC. The elimination of HCI during thermal degradation of PVC leads to the formation of olefinic double bonds in the pol ymer molecule. Therefore, provided that the extent of degradation is low, and dehydrochlorination is not accompanied by secondary reactions involving the double bonds, measurement of the level of unsaturation in the degraded polymer sample should provide satisfactory estimate of the relative stabilising ability of additives. The ratio of the iodine value of degraded PVC sample to the value of undegraded sample, IV/IV0 , can, therefore, be used to assess the relative
566 J SCI INO RES VOL 59 JULY 2000
effect of the tubber seed oil derivatives on the thermal degradation of PVC: the smalle r the IV/IV0 value the greater the re lative stabili sing ability of the additive. The results in Table 3 show that the additi on of meta l soaps o f tubber seed oil to PVC reduced the level of unsaturation in the degraded polymer sample.
Thermoxidative Degradation
The viscos ity paramete rs and levels of
unsaturation of PVC samples degraded in a ir at 190°C in the presence of metal soaps of tubber seed o il are shown in Table 4. The results show that the values of
intrinsic viscosity of the degraded PVC samples are lower than the value for the undegraded sample. It can be seen from Table 4 that whereas the va lue of intrin sic viscos ity of the PVC sample degraded without the additives is onl y about 30 per cent of the value of the undegraded sample, the values of intrinsic viscos ity of PVC samples degraded in the presence of meta l soaps of rubber seed o il are much higher with values of between 60 and 90 per cent of the va lue of the undegraded sample. These results indicate that the metal ·soaps of rubber seed o il exe11 stabilising effect on the ox idati ve thermal degradation or PVC. T he Huggi ns' constant is genera ll y used as a measure of polymer-solvent and polymer-po lymer interacti on. In conjunction with intrinsic viscosity va lues the Huggins ' constan t, be ing a parameter indicative of variations in intermo lecul ar forces, provides an overa ll view of macromolecular conformation and behaviour
in solution.
Table 3- Level s of unsaturati on in PVC samp les degraded at
180°C in the presence of rubber seed oi l der ivatives for 30 min .
Additi ves Iodi ne Rel ative
value (I V) iodine value
(gl"/ 1 OOg) (IV)/(IV).,
None 61.76 6.50
RSO ( I 0 wt per cent ) 55 .24 5.81
ERSO ( I 0 wt per cent)
4.8 per cent epox ide conten t 58. 17 6.12
I 0.6 per cent cpox idc content 48 .47 5.10
Meta l soaps of RSO (3 wt per cent)
Ba riu m soap 38 .77 4.08
Cadmium soap 35 .55 3.74
Lead soap 29.08 3.06
The iodine value obtai ned for PVC samp les degraded in the presence of tubber seed o il deri vatives are marked ly lower, by as much as 85 iodi ne value units, that the corresponding va lue for lhe PVC sample degraded in the absence of the add iti ves. It can be seen that re lati ve iod ine value, IV /T\/ 0 , var ied from abou t 18.0 for PVC sample degraded in the absence of ru bber seed oil additi ves to about 9.0 fo r PVC sample degraded in the presence of cadmiu m soap of rubber seed oil. These results indicate that the rubber seed o il de ri vati ves stab ili se PVC aga inst ox idative thermal degradati on.
Table 5 shows the viscosi ty parameter for PVC samples degraded in the presence of meta l soaps of
Tabl e 4- Thermoox idativc degradation of PVC at 190°C in the presence o f metal soaps of RSO. Intrinsic viscos ity and levels of
un saturation in the degraded polymer samp les.
Additive Peri od of Huggi ns' Intrinsic Relat ive intrin sic Iodine va lue (IV ) Relative
degradati on constan t viscosity [111 viscosity g i/ IOOg iod ine value
(min) K,, (d l/g) 11111111 L sample (IV/IV),.
None 30 23. 14 0.59 0.3 1 174.63 17.83
60 17.64 0'.60 0. 32 180.20 18.39
Barium soap of RSO 30 7. 12 1.1 4 0 .60 110.20 11.25
60 9.94 0.88 0.46 11 9.75 12.22
Cadmium soap of RSO 30 2.03 1.70 0.89 88.93 9.08
60 3.80 1.50 0.80 89.90 9.17
Lead soap of RSO 30 2.84 1.52 0.80 100.06 10.2 1
60 4. 13 1.34 0.71 100. 34 10.24
OKIEIMEN: THERMAL STABILISATION OF POLYVINYLCHLORIDE 567
Table 5- Viscosity parameters ror PVC degraded at 190°C under nitrogen and in the presence or epoxidised Jatropha seed oil (EJS O)
and epoxidi sed rubber seed o il (ERSO) and their metal soaps.
Add itive Period or r 111 Kll [ 11]/[T)].,
degradation (min) (d llg) Huggins' constant
None 30 0.46 39.50 0.4 1
60 0.37 - 42.78 0.33
EJSO ( I 0 wt per cent )
5.6 per cent epoxide content 30 0.77(0.7 1) 17.90 (2 1.48) 0.69 (0.64)
60 0.55 (0.43) 27 .84 (54.08) 0.49 (0.39)
9.5 per cent epox ide con tent 30 0.83 I 6.33 0.75
60 0.61 26.87 0.55
ERSO ( I 0 wt per cent)
4.8 per cent epoxide contelll 30 0.72 (0.67) 14.42 ( 18 80) 0.65 (0.60)
60 0.41 (0.40) 38.65 (42.25) 0.37 (0.36)
I 0.6 per cent epox ide content 30 0.81 12.35 0.73
60 0.52 25.07 0.47
Metal soaps or EJSO (9.5 per cent epoxide content) (3 wt per cent)
Barium soap 30 0.80: 0.76" 18.23 : 17.3 1" 0.72: 0.68"
60 0.58: 0.62" 29.73: 24.7 1" 0.52: 0.56"
Cadmium soap 30 1.02: 1.0 1" 9.62: I 0.00" 0.92: 0.91"
60 0.76: 0.7 1" i7 .3 1: 19.84" 0.68: 0.64"
Lead soap 30 0.98: 0.92" I 0.42: I 1.8 1" 0.88 : 0.83"
60 0.69: 0.58" 21.00: 33.03" 0.62: 0.52"
Zinc soap 30 0.98 : 0.88" 10.42 : 12.91 " 0.88: 0.79"
60 0.62: 0.58" 24.28: 37. 16" 0.56: 0.52"
Metal soaps or ERSO ( I 0.6 per cent) epoxide content ) (3 wt per cen t)
Barium soap 30 0.90: 0.80" 8. 11: 12.59" 0.81: 0.72"
60 0.64: 0.64" 16.88: 16.88" 0.58: 0.58"
Cadmium soap 30 0.90: 0.69" 8.64: 17.85" 0.8 1: 0.62"
60 ()59: 0.57" 21.63: 21.88" 0.53: 0.51 "
Lead soap 30 0.99: 1.00" 7.02: 6.88" 0.89: 0.90"
60 0.72: 0.68" 13.27: 14.78" 0.65 : 0.61 "
{I)] and K values of PVC samples degraded in the presence of 1111epoxidised oils in parentheses, { 1)/, = l'(r/ue of intrinsic l'iJCosin· of
undegraded PVC sample, a= 1•alue of {17{ and K obtainedfor PVC salllflles deg raded in the pre.mrce of metal soaps of wrqwxidiscd oils.
epox idised rubber seed oil (ERSO), and epox idi se Jatropha seed oil (EJSO). It can be seen that at comparab le levels of epoxidat ion and in the presence of simil ar metal soaps of the epox idised o il s the va lues of [T)]/[T)]n obtained with PVC samples degraded with Jatropha seed oil derivatives are higher than the values obtained for PVC degraded with rubber seed o il derivatives. These results indicate that the derivatives of Jatropha seed o il are relatively more effecti ve in stabi li si ng PVC against thermal degradation than derivatives of rubber seed oi l. It has been suggested that the nature of the hydrocarbon portion of the fatty acid component of vegetable oil s may influence PVC
stab ility by a lteri ng the polymer-matrix polarity: 2
a lthough the exact nature of the effect is not fully uncjerstood. In a previous stud/ 7 it was reported that metal soaps of rubber seed oil were more effect ive than the corresponding soaps of unsaturated fatty acids (o le ic and linoleic ac ids) in stab ili sing PVC aga inst thermal degradati on and explanations in terms of the contribution of the chemical identity of the hydrocarbon moiety of the soaps, to the overall stabilisation process were proffered. The relati ve ly high level of C 18 unsaturated fatty acid . conten t of rubber seed o il (64.23 per cent) may account for the observed lower stabili sing ab ility of de ri vati ves of
568 J SCI IND RES VOL 59 JULY 2000
rubber seed oil. The unsatu rated fatty content of Jatropha seed oi l is 41.71 per cent28 while that of rubber seed oil is about 80 per cent.
Conclusion
This study shows the relative effectiveness of rubber seed oi l derivatives a~ thermal stabi li ser for PVC. The results indicate a potential for the appl ication of epoxidised rubber seed oi l and its metal soaps as thermal stab ili sers in PVC fo rmulations.
References Guyot A, Roux P & Tho P Q, Sur ,,\ Degradation Thcrmique du Polychlorure du Vinyle (IV) II lluence de Ia Rcgu laritc Structurale des Polymeres sur Leur )tabi lite Thermique Sous Atmosphere lnerte, J Appl Polym Sc '· 9 ( 1965) 1823 - 1840.
2 Geddes W C, Thermal Degradati01 and Polyvinyl ch loride, Rubber Chem Tech, 40 (1967) 177 - Z48.
3 Varma I K & Sharma K K, n erma! Degrad ation of Polyvinylchloride in the Presence .,f Additives (Part I), Angew Makromol Chem, 71 ( 1978) I 57- 166.
4 lida T & Goto K, Stabi li sation of Polyvinylchloride (v). Synergism Between Metal Soaps and Po lyols upon Stabilisation of Polyvinylch loride, J Appl Polym Sci, 25 ( 1980) 887 - 900.
5 Okicimen F E, Stmctural Defects and Thermal Stability of Polyvinylchloride, Bull Chem Soc Niger, 6 ( 198 1) 50 - 58.
6 Lewis J, Okieimen F E & Park G S, Stabi li sing Effect of Allylchloride Comonomer Units on the Thermal Degradat ion of Polyvinylchloride, J Macromol Sci (Chem). A17(6) ( 1982) 1021 - 1023.
7 Wypych J , Polyvinylchloride Degradation (Elsevier, Amsterdam) 1985.
8 Varma I K, Ohara S K, Varma M & Biddapa T S, Cashew Nut Shell Liquid: Characterisation and Studies of PVC Degradation, Angew Makromol Chem, 54 (1987) 67-86.
9 Simon P, Kineti cs of Polymer Degradation Invo lving the Splitting Off of Small Molecules (Part I). Basic Concepts, Polym De grad Stabilis, 29 ( 1990) I 55- 163.
I 0 Simon P & ·Valko L, Kinetics of Polymer Degradation In volving the Spli lling orr of Small Molecules (Part 2). Thermal Degradation or Polyvinylchloride in Inert Atmosphere, Polym Degrad Stabilis, 29 ( 1990) 253- 261.
II Simon P, Gatial A & Valko L, Kinetics of Polymer Degradation Involving the Splitting OtT of Small Molecules Part 3. Treatment of Experimental Data for Dehydrochlorination of Polyvinylchloride in An Inert Atmosphere, Polym De grad Stabilis, 29 ( 1990) 263 - 270.
12 Starnes W H (Jr ), The Frye-Horst Mechanism for the Stabilisation of Polyvinylchloride by Organic Metal Salts; A Critical Assessment , Polym ?reprint Am Chem Soc, Poh•m
Sci, 35(1) (1994) 425-426.
13 Bacaloglu R & Fisch M, Degradation and Stabi lisation of Polyvinylchloride (1 ), Kineti cs of Thermal Degradation of· Polyvinylchloride. Polym Degrad Stabilis, 45 ( 1993) 30 I -3 13.
14 Bacaloglu R & Fisch M, Degradation and Stabilisation of Polyvinylchloride (V), Reaction Mechanism for Polyvinylchloride Degradation, Polym Degrab Stabilis, 47 ( 1995 ) 33-57.
I 5 Abbas K.B & Sorvik E M, On the Thermal Degradation of Polyvinylchloride (IV). Initi at ion Sites of Dehydrochlorinati on, J Appl Polym Sci, 20 ( 1976) 2395 - 2406.
16 Kl emchu k P P, Po lyviny lchloride Stabilisation Mechanisms, in Stabilisation of Polymers and SJabiliser Processes, Advances in Chemistry Series 85, edited by R F Gould , American Chemical Society. Washi ngton DC) Chap I, 1968 , pp I - 17.
17 Frye A H & Horst R W, The Mechan ism of Polyvinylchl oride Stabi lisation by Barium, Cadmium and Zinc Carboxylates, J Polym Sci, 40 ( 1959) 419 - 431 .
18 Hopff H, Polyepoxides as Stabiliser fo r Po lyvinylchloridc in Stab ilisation of Polymers and Stabiliser Processes, Advances in Chemistry Series 85, edited by R F Gould (American Chemical Society Washington DC) Chap 5, ( 1968), pp 57 -64.
19 Epoxy Resins. Chemist!)' and Technology, edited by C A May, Second ed (Marcel Dekker Inc, New York ) 1988, pp.947.
20 Aigbodion A I, Studies in the Utilisation of Rubber Seed Oil in the Preparation of Alkyd Resins, Ph D Thesis, University of Benin, Benin City ( 1995).
21 Durbetaki A I, Direct Titration of Oxirane Oxygen with Hydrogen Bromide in Acetic Acid, Anal Chem, 28 ( 1956) 2000- 2004.
22 Aigbodion A I, Okieimen F E & Bakare 0 I, In Situ Epoxidation of Rubber Seed Oil by Peracetic Acid, Niger J Appl Sci, (1999) 21-32.
23 Burrows H D, Ellis H A & Akanni M S, Proc Second Eur
Symp on Thermal Analysis, Heyden, London ( 1981) pp 302-305.
24 Odi lora C A, On the Thermal Degradat ion of Polyvinylch loride, Synergistic Effect of Mixtures of Lead and Zinc Carboxylates, Acta Polym, 40 ( 1989) 54 - 545.
25 BSI, Briti sh Standard Institution, Methods of Analysis of Fats and Oils RS 684 ( 1976).
26 Minsker K S, Abdullin M I & Zaikov G E, Liquid-Phase Degradat ion of Polyvinylch loride, J Vinyl Tee/mol, 34 ( 1981) 230-239.
27 Gkieimen F E, Utilisation of Rubber Seed Oil in the Stabilisation or Polyvinylchloride Against Thermal Degradation, Niger J Chem Res, ( 1999) (in press).
28 Sogbaike C E, Studies in the Thermal Degradation of Polyvinylchloride , Ph D Thesis, Universi £y of Benin , Benin City (1995).