'l' P '1

III :

ellf:. (1) t. I

11'1

' Illl f, 11

Il c: If,

lit f'

/I I\:

I·r. tI/

·i· c 6 ..

he en

a ~e

d e

PALI-SK 69-0827

Palit & Tripath)': SYllthesis of Benzoic Acid from Clt/oroh.en::el1C' IIl1def Pressure 301

References

" iloski Aoki, J. clectrachell/. Soc. Japan , 1967, 35, 137 ; Warne, ,,1. A., & Hayfield, P. C. S., Trans. fllSt. Metal Finish., 1967, -'5,83

,\damson, A. F., Lever, B. G ., & Stanes, 'N. F., J. appl. Chem., Lonc/., 1963, 13,483

, Greeson. H.E., to be published • c\ yres, G. H., & Meyer, A. S., Analyt. Chem., 1951, 23, 299 • 11> 1. N., & Landolt , D., J. £'Iectmchelll. Soc., 1968,115,713 ' Antler, M., & Butler, C. A., Electrachem. Technol., 1967,5, 126

7 I-I oare, J. Poo 'The Electrochemistry of Oxygen', 1968 (New York: Intersciencc)

8 Anson, F. c., & Lingane, J. J., J. Am. chern. Soc., 1957, 79, 4901

• Breiter, M. W., & ¥leininger, J. L., J. electrochem. Soc., 1962, 109, 1135

10 Suzuk, Coo Yoshida, Moo Onoue, H., & Matsumo, T., J. electro­chem. Soc. japan, 1966.34, 165

II Flisskii, M. M., Electrokhimiya, 1966,2,860

SYNTHESIS OF BENZOIC ACID FROM CHLOROBENZENE UNDER PRESSURE

By S. K. PALIT and N. TRlPATHY*

Conversions as high as 76'26 % and 80 ·85 % of chlorobenzene to benzoic acid were obtained when chlorobenzene was carboxylated with ca rbon monoxide and water at high pressure with nickel iodide-silica gel catalyst (Ni:Si0 2

= 50: 50) and without any ca talyst (thermal) respectively, at the opt imum temperature and pressure; these optima were much higher for the latter than for the former process. The lower yield in the catalytic process could be improved to some extent if the catalytic reaction was carried out at a pressure, reached by introdlicing nitrogen in addition to the usual input of carbon monoxide, almost equal to the optimum pressure of the thermal process.

Introduction

The possibility of synthesising various aromatic carboxylic acids, esters etc. of industrial importance by the carbonyla­tion (or carboxylation) of aryl halides under pressure with mainly nickel carbonyl as catalyst, has been disclosed in various patents.l In one such patent, Kroper et al. 2 have claimed that benzoic acid and its esters can be synthesised from bromo- and chloro-benzenes in a similar way. Romanov­skii & Artemev3 in a paper on carbonylation of dichloro­benzene, have also claimed to have obtained benzoic acid by the same procedure. However, as both these papers are lacking in detail and because of the industrial importance of

benzoic acid and the superior catalytic activIty of nickel iodide-silica gel catalyst (Ni : Si02 = 50 ; 50) to the car­bonyls,4 a detailed study of the synthesis of benzoic acid from chlorobenzene, carbon monoxide and water with and without this catalyst has been carried out.

Thermodynamics of the reaction

The effects of temperature and pressure on the equilibrium constant of this reaction are given in Table I.

TABLt I Equilibrium constant of the reaction (a) at mrious temperatures and a pressure of 1 atm. and (b) at various pressures

for a constant tempera ture 200 'c Standard free energy change for the' reaction at 25"c and 1 atm = -15-83 kcal

(a) Temperature, °c 100 150 200 250 300 350 400 450 K "p· 8'l3x107 1·51 x 10' 1·82xI0' 4-78 x 10J 5·88 X 10' 9·77 x 10 2'31 x 10 5-91

(b) Pressure, atm. 100 200 300 400 500 600 700 K* 0·7789 1·236 1·516 2·607 3·716 3-627 3·621 K'p· x 10- 5 2·34 1'47 1·201 0-698 0 '-l99 0-501 0·503

·K', = KO •• where KOp = equilibrium constant at I atm; K'p = equilibrium constant for diO"erent pressures at Kv

constant temperature and K,. = YC f, .. , r OOIl X '(IICI and Y = _pi where I is the fugacity of the gas and y is the YC6I"CI X Yeo X Y .. ,o

·Present addre s: Dept. of Chemistry, Regional Engineering College, Rourkela, Orissa. India.

J. appl. Chern .. 1969, Vol. 19, October

302 Palil & Tril'olhy: SYlllhesi.\· of Bell::oic Acid from Chlorohel1::el1c Illlder Pressure

As values of the free energy and heat of formation of benzoic :ll:iti in the gaseous sta te anti the heat capacities of benzoic a ' id and chlorobenzene \\ere not availabh:, these were cal­culated by the group contribution method. s.

For the calculation of the clrcct of pressure 011 the equi­librium con tant at 200 ', the fugacity co-etllcients of the various compounds were evaluated -from the chart proposed by Lydersen ('/ al. 6 and corrected for the dilTercnt critical com pre. si bilities. 7 The temperature of 200-' was selected so as to avoid the unreliable reduced temperature range of 0,8-1·2 at reduced pressures above I '2.7 From Table I(b), it is seen that the equilibrium constant contrary to expectalion, fluctuated in value. This could be due to the influence of pressure on the fugacity co-enicients of the various com­pounds, but as the reaction proceeds with volume decrease it is obvious that the equilibrium conversion would increase with pressure.

Experimental

A high-pressure rocking autoclave of 270 ml capacity at 25"c and I atm pressure having an angular play of 15° with 45 oscillations per min was used. The product release assembly and other experimental set-up were similar to those des­cribed by Bhattacharyya and co-workers. 8 ,9

Very pure carbon monoxide was prepared in the laboratory by the same method and the same precautions were taken as in the earlier work. 8 ,9 All other chemicals used were of analytical grade.

The catalyst, nickel iodide supported on silica gel con­taining 84'16 % of nickel iodide (Ni : Si02 = 50 : 50), used in this work, was prepared according to the method of Bhattacharyya & Sourirajan.9

Liquid products The liquid products of the reaction consisted of benzoic

acid, formic acid and benzaldehyde (in a few cases under non-catalytic conditions only), together with the unconverted chlorobenzene and watef'. Occasionally minute traces of hydro hloric acid, nickel and iron salts were also present. lndi\ idual components were identified by their melting points and mixed melting points wherever possible, by forming suitable derivatives and 'by paper and gas chromato­graphy.

The liquid product mixture was cooled to abou t 0° and filtered. The residue, containing benzoic acid and other solid impurities, was treated with alcohol and subsequently titrated with standard alkali to estimate benzoic acid. A quantity of the filtrate was steam-distilled to eliminate the impUrities. In the distillate, benzaldehyde was estimated by the 2: 4 dinitrophenyl hydrazine method; formic acid by titration with alkaline potassium permanganate and then subtracting the amount consumed by benzaldehyde (if present) from the above titration value; and finally benzoic acid was determined by subtracting the acidity due to formic acid from the total free acidity of the distillate. As the amount of hydrochloric acid or the chlorides, determined by Volhard's method after acidifying a part of the filtrate with nitric acid, was almost negligible, it did not interfere with the analyses of the other products.

Gaseous products The gaseous product was found to contain carbon dioxide,

hydrogen and saturated hydrocarbons (methane was the chief con titucnt) together with carbon monoxide. Analysis was carried out with a standard Orsat gas analyser and the

results ,,"ere confirmed when nece sary by gas chromato­graphy.

Results and disclission

As some preliminary iI1\estigations revealed that benZOl, acid was produced even when the experiments were can, ducted in the absence of any catalyst (thermal), the uek-r. mination of the optimum reaction conditions at which thl' yield of benLoic acid would be a maximum was carried 0 111

for both the catalytic and non-catalytic conditions. The result, are recorded in Figs 1--4 and Tables ll- JV. Since the P:lr. ticular reactions by which carbon dioxide and saturated hyd rocarbons were formed during the synthesis reaction, could not be represented by balanced equations, the yields of these gases were calculated on the ba is of the percentage of total input carbon (from chlorobenzene and carbon mono oxide) that could be accounted for by the carbon of th~ respective gas and have been given as 'percentage carbon conversion'. Similarly the yield of gaseous hydrogen has been expressed as 'percentage hydrogen conversion.'

Effeet of presslire The influence of pressure at different temperatures \.\ a,

studied under both catalytic and non-catalytic condition, by varying the input of all the reactants by the same pro· portion so that their mole ratio remained constant in all the: experiments (Fig. 1). When the temperature was changcd by 90° (220- 3) 0°) in experiments with catalyst, the opti· mum pressure did not increase to the extent that was notcd by only a 50° rise (310--360°) under non-catalytic conditions. This was probably because the temperature (360°) was vcry close to the critical temperatures of water and chloro· benzene. With increasing pressures the yields of benzaltk· hyde, formic acid and the gaseous decomposition product '; increased steadily in all cases. However under catalyt ic conditions the yields of formic acid and gaseous decompo­sition products were higher than those under non-catalytic conditions (Fig. 2).

Effect of temperature The optimum temperature of 295° under catalytic con­

ditions was much lower than 360~ for non-catalytic COil'

ditions (Fig. 3). r n both cases, the yields of benzaldch) lk and gaseous decomposition products increased with rise r temperature whereas the yield of formic acid decreased. Of the decomposi tion products the yields of carbon dioxide, sa'lUrated hydrocarbon and hydrogen, increased steadily fro l11 12 ·9--42'8%, 4 ·7- 6 ·7 % and 0·5-30·3 % respectively for catalytic conditions, whereas for the non-catalytic condition,. the corresponding values were respectively 19 ·2-31 ·2",. 0- 11·6 % and 0- 11,6 % over the temperature ranges inve,lI­gated.

EtTeet of residence time With increase in residence time the yield of benzoic a.:id

passed through an optimum at three hours in both ca~c~ . However in the catalytic synthesis the yield of benzoic ali ll

was Jess affected by residence time than when under non· catalytic conditions. The yicId of formic acid increased ;1'

residence time wa increa ed for catalytic synthesis \\llIk the reverse was true for non-ca talytic synthesis. The yield, .,1 gaseous decomposition products increased steadily \\l lh residence time and their yields increased almost by the sail" amounts in the two different syn thetic conditions 1'01' tI l<' same variation in residence lime.

J. appI. Chent., 1969, Vol. 19, Oc(()hl'f

-t­a u < u o N Z UJ CD

u. o a ...J UJ ;;:

Fig. 1.

With

Fig. 3

Me

In

CO

2·33 2' 15 1·97

~ J. :pl

n •

(', ...

",. . h t

Id . \ "U .

t' I' . ,It ,ll ,

(I I, .. "

lu\ " apt, (-, m,l:,

If II, rbI" h.,

\I ,1\

li(llh

fin'· I thl' ngl'd vpll · Ot~d

ion,. \ er~

oro-Ide·

ron­;on­Iyde e of

Of ide. om for

'ns, 10 / . .,. ,'. :sl i-

cid es. -id n-as ik of 'Ih lC

hI!

rr

... ·teem

..

Palit & TripathJ' ,' Synthesis oj Ben::oic A tid frOI/l ChlorobeJI::ene linder Pressure 303

f-0 u 4:

u 0 N Z W cD

u. 0

0 ..J w ;;

BO

40

O~ ____________ ~ ____________ -L ____ ---'

o 4000

PRESSURE,lb/in'

8000

Fig. I. Eifect of pressure all yields of bellzoic acid ill both catalytic and thermal processes

With catalyst: _O---O 5 mlat 31O' c; .-----. 10 ml at 220°c :-Jon-calalytic: 0--0 at 360°c; 11-----. at 310°c ~lole ratio of reacta nts: CO : H 20: C.H,C1 = 1 : 1·29: 0 ·0114 Residence time = 3 h

o 0-o u < u o N

BO

~40 cD

u. o o ...J UJ

>

O~------------'L-------------' 200 300

TEMPERATURE,oc

400'

Fig. 3. Effect of tempera lure 01/ yields of bel/zoic acid ill both catalytic alld thermal processes

Mole ralio of reactants CO:I-I,O:C.H,C1 = 2·15:2·77:0·0246 With catalyst: 0 --0 5 ml a t 5700- 9'+00 Ib/ in.> Non-catalytic: 0--0 at 4000-9600 Ib/ in.'

30 :; Q. c:

a 0

0-vi I-

g2 0 0 Q: Il. (/)

=> 0 w (/)

< ~1O u. 0 0 ...J UJ

>=

",,,,..,...,...,...,. _-cr-------o

0"'"

--- ..... ---- ... ..,. .... -- /

,.."," ~

.."" 0----- /

___ ~'--/-----6 - - lr=·!:==-=::'=="-~

•• ------~*r-------~.~~ O~-------------J--------------~------~ 4000 6000

PRESSURE,lb/in'

BODO

Fig. 2. Eifect of pressure on yields of gaseous decolI/positioll pro­ducls al 310 ' c ill bOlh catalytic {Jlld therll/al processes

With catalyst: 0 --0 CO 2 ; 0--0 H 2 ; 6,--6, saturated hydrocarbon

Non-catalytic: .--. CO 2 ; 11--. H , ; 4.--,A. saturated hydrocarbon

Reaction conditions are same as in Fig. 1

'!­£F -u < u 80 o N Z UJ cD

u. o o ...J UJ

>=

RESIDENCE TIME, h

Fig. 4. Eflect of residence time all yields of bell=oic acid ill b011r calalytic alld tlrerma! processes

Reacta nts ratio as in Fig. 3 With catalyst: 0 -----0 10 ml at 295°c and 6700Ib/ in.> Non-catalytic: 0--0 at 360 c and 9 100 Ib/ in.'

TABLE ](

Effect of ,-ar ia lion of reactant ratio

Percen tage of inrllt Input of reactant. Yield of Percentage of carbon converted to hydrogen converted

moles Press.rre. Ib/ in 1 benzoic acid, 0 / CO, Sat. hydrocarbon to gaseolls hyd rogen / 0

CO 11 ,0 C.H,C1 Catalytic Thcrmal Catalytic Thermal Catalytic Therma l Cata lytic Thermal Catalytic Thermal

2·33 1'66 0 ·0246 6500 8200 4 1·2 1 64-47 18 ·7 17·9 3·0 7-9 23 ·0 7 ·5 2' 15 2·77 0'0:!-l6 6750 9100 74·94 RO 'R5 23-8 27·5 ·9 '3 9 '8 9·2 10·5 1·97 3-89 0 ·0246 6900 9150 66 ·6 1 5-+-95 26 ·30 30·8 10-4 12·3 6·0 7·0 2· 15 2·77 O'M92 !!800 67 ·11 19 ·9 6 ·0 9 ·6 2·M 3·33 0·0-+92 6700 3 1' 17 21·8 3·7 11 ,8

Reac tion conditions: catalytic- with 101111 nic l..cI iodide:sil ica gel cata lys t at 295"c for 3 h ; thermal (no n-catalytic)- at .160 c for 3 h.

J. ~ppl. ChCIll., 1969, Vol. 19, October

• 304 Palit & Tripath.\': Sy!!ti1esis (If BCI1::oic Acid/rom Ch/oroben;:cl1c under Pr('.uure

Effect of changc in reactant ratio As is e\'ident frol11 Table II, the optimum yield for both

eatalyti\.: and non-catalytic syntheses was obtained when the input of carbon monoxide, water and chlorobenzene were 2'15,2'7 and 0 ·0246 moles respect ive ly. With increase in the input and mole ratio of water and carbon monoxide, the yields of c3rbon dioxide and methaile increased steadi ly for both conditions whereas hyd rogen did not foll ow any par­ticular trend. With increasi ng amounts of water the yield of formic acid increased for catalytic reactions whereas for the non-ca talytic reaction there was little change.

Elfect of the quantity of catalyst It is c\'idcnt from Table III that an increase in the amount

of catalyst beyond 10 ml (13 g) did not have any appreciable effcct on the yield of benzoic acid or gaseous decomposition products. The yield of formic acid (not presented in Table III) was highest with 15 1111 of the catalyst, which incidentally also gave the optimum yields of benzoic acid.

TABLE III Effect of catalyst amount

Catalyst volume, ml 5 10 15 20

Yield of benzoic acid, % 72·26 74·94 76·26 75 ·89 Amount of carbon converted, %

(i) to CO 2 23·7 23·8 24 ·4 25-8 (ii) to saturated hydrocarbon 4·8 9·3 9·4 9·5

Amount of hydrogen converted to gaseous hydrogen, % 2·8 9·2 9·2 9·5

Reaction condiiions: Pressure 67501b/in 2 at 295°c for residence period of 311; mole ratio of reactants CO; H 20 : C6 H,C1 = 2·15 : 2·77 : 0 '0246; 10 ml of catalyst weighs 13 g.

Conclusions

In the systems studied it was found that nicke l iodide lin silica gel (Ni : SiOl = 50 : 50) is an cOicient catalyst gi\ in ~ a much lower optimum reaction pre sure and temperat urt:. 67501b/ in 2 and 295- respectively, compared with the vulut" 9100 Ib/ in 2 and 360 ' re pectively, under non cntalytic Con. diti ons (Table I V) with only about 4 % loss in the ben7ol. acid yield.

When the optimum inpu\ pressu re of 2500 Ib/i n2 at 25 in the catalytic study was boosted with nitrogen to an input pressure of 35001b/ in 2 at 25° so that the final reaction pressure of 9700 Jb/ in2 was almost identic31 with the opti. mum pressure under non-catalytic conditions, the percenta l!t' con ersion of chlorobenzene to b;:nzoic acid increased til nearly the same va lue as th.1t under non-catalytic condition (Table IV), indicating the pressure dependence of the reaction

Production of formic acid, probably by the interaction (It

carbon monoxide with water, was favoured to a grcah:r extent in the presence of the cataly t. Similarly decompo. sition or other side reactions leading to the formation of various gaseous products, were more favoured in presence of the ca talyst.

Acknowledgment

The authors wish to thank Prof. S. K. Bhattacharyya for his helpful suggestions and keen interest in the work.

Department of Applied Chemistry, Indian Institute of Technology,

Kharagpur, India

Received 9 May, 1909

TABLE lV Comparative yields under optimum conditions for the catalytic and thermal processes

Reaction temperature, °c Input pressure, Jb/in' Reaction pressure, Ib/in 1

Yield of benzoic acid, % Amount of carbon converted to:

(i) CO2 , % (ii) saturated hydrocarbon, %

Amount of hydrogen converted to gaseous hydrogen, %

Thermal

360 2500 9100

80·85

26·1 10·3 10·2

Catalytic (reaction pressure Catalytic raised by nitrogen)

295 295 2500 3500· 6750 9700

76·26 79 ·14

24·4 17·8 9 ·4 0 ·7 9·2 0·2

*3500 Ib/ in 2 initial pressure was built up by an input of 2500 Ib/in 2 carbon monoxide and 1000 Ib/in' nitrogen. Reaction conditions: mole ratio of reactants and residence time as in Table Ill ; volume of catalyst 15 ml.

Referenccs

I Bird, C. W. , 'Transition Metal Intermediates in Organic Syn­thesis', 1967, p. 192 (London: Logos Press & Elek Books)

2 Kroper, H., Wirth. F., & Huckler, O. H ., Ger. P. 1,074,028; Chelll. Abslr., 1961,55, 12364f

3 Romanovski i, V. I. , & Artemev, A. A., Zh. vses. khilll. Obshch., 1960,5.476; Chem. Abslr., 1961,55, 1506 d, 3511g

" Bhallacharyya, S. K., J. iledian deelll. Sac., 1968, 45, 1151 , Hougen, O. A., Watson, K . M., & Ragatz, R. A. , 'Chemical

Process Principles', 2nd Edn, 1959, Part 11, p. 100.:1 (New York : John Wiley)

• Hougen, O. A., V';alson, K. M., & Ragatz. R. A. , 'Chcll1i, .li Process Principles Charts', 2nd Edn, 1960, Figs ]42:1, I·C I. (New York: John Wiley) .

7 Hougen, O. A., Watson, K . M., & Ragatz, R. A., 'ChcI111.:.11

Process Principles', 2nd Edn, 1959, Part II, p. 574 ( ,'\' York: John Wiley)

8 BhatLacharyya, S. K ., & Sen, A. K., J. appl. Chelll., LOlld. 1963, 13, 498

• Bhattncharyya, S. K., & Sourirajan, S., J. appl. Clcelll., LIIII.I. 1956,6,442

J. appl. Chcm., 1969, Vol. 19, Odobt·r