The Reactivity and Selectivity of the Reaction of Sulfur Trioxide and Bromobenzene

Brigham Young University BYU ScholarsArchive

Theses and Dissertations

1970-06-01

The reactivity and selectivity of the reaction of sulfur trioxide and bromobenzene

Sullivan E. Blau Brigham Young University - Provo

Follow this and additional works at: https://scholarsarchive.byu.edu/etd

BYU ScholarsArchive Citation Blau, Sullivan E., "The reactivity and selectivity of the reaction of sulfur trioxide and bromobenzene" (1970). Theses and Dissertations. 8171. https://scholarsarchive.byu.edu/etd/8171

This Dissertation is brought to you for free and open access by BYU ScholarsArchive. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of BYU ScholarsArchive. For more information, please contact [email protected], [email protected]. &.V 1-022-- .. 85? f'J?O

THE REltC'I 1IVITY A..1\10SELECTIVITY OF THE REACTION OF

SULFUR TRIOXIDE AND BROMOBENZENB

A Dissertation Presen·t:.ed to the Department of Cha~istry Brigham Young University \

In Partia.l Fulfillment of the Requirements tor the Degree Doctor of ~hilosophy

by Sullivan E. Blau

J·une 1970 This dissertation, by Sullivan E. Blau, i.s accepted in its present form by the DepaS"tmeht of Chemistry of Brigham Young University as satisfying the dissertation requirement for the degree of Doctor of Philosophy.

ii ACKNOWLEDGMENTS

I would like to express appreciation to the staff and faculty in the Department of Chemis·try, Brigham Young University, for cont.inued assistance and support during my tenure here. Dr. K. LeRoi Nelson is due special thanks for his interest, assistance and understanding extended to me in spite of many other responsibilities and demands for his time and consideration.

__,_

iii TABLE OF CONTENTS

ACKNOWLEDGMENTS• • • • • • • • • • • • • • • • • • • • iii LIST OF TABLES• •• • • • • • • • • • • • • • • • • • • vi LIST OF ILLUSTRATIONS • • • • • • • • • • • • • • • • .viii

Chapt,;:r I. INTRODUCTIONAND BACKGROUND • • • • • • • • • • l Electrophilic P..romatic Substitution Reactivity and Selectivity Electrophilic Aromatic Substitution Reactions on Bromobenzene

II. AROt-iATIC STJLFO:NATION • • • • • • • • • • • * • • 19 Secondary Reactions Secondary Reactions and Their Effects on Relative Rates Secondary Reactions and Their Effects on Isomer Distribution Kinetic Isotope Effects Reversibility III. RESULTS AND DISCUSSION. • ••••• • • •• • • 39 Sul.fodehalogenation Calculation of Relative Rates 'l'est for 'X'hermodynamic vs. Kinetic Control Analysis of the Data for Secondary Reactions Isomer Distribution for the sulfonation of Bromobenzene Calculation of the Partial Rate Factors Error Ana.lysis

iv Chapter Page

lV. EXPERIMENTALSECTION • • fj • • • • • • • • • • • 71 Preparation of the' Isomers of Bromo- benzenesulfonic Acid Salts Preparation of Radioactive sulfur Trioxide Preparation of Radioactive Benzene Purification of Benzene and Bromobenzene Sulfonation Procedure Analysis of the Isomer Distribution EXperiments Analysis of Competitive sulfonation EXperiments

LIST OF REFER~CES • •••• • • • • • •• • • • • • • • 86 APPENDIXES

A. Generation of Data t.o Test the Effects of secondary Reactions ...... 94

B. Design of a Flow Apparatus for the Study of the Bromination of Olefins • • • • • • • • • • J.C.l2 c. A Research Proposal, A Study cf the Stability and Reactions of the Cy,.:!lopentenyl Free Radical ••••••••••••••••• lll D. Manuscript for Publication: Reactivity and selectivity of t'he Reaction of sulfu.t Trioxide and Bro:mobenzene •••••••••• 117

. V LIST OF TABLES

Table Page 1. Correlation of Halogenation Rates with Complex Stabilities • • • • • • • • • • • • • • • • • • 8

2. Selectivity and Reactivity of Electrophilic Substitutions on Toluene ••••••••• • • • 10 3. Partial Rate Factors for Electrophilic Substitutions on Bromobenzene ••• • • • • • • 14 4. Reactions Having Large Deviations from the Selectivity Relationship ••••••••• • • • 16 s. Isomerization and Desulfonation of Toluene- sulfonic Acids •••••••••••••• • • • 22 6. Data Generated with a Primary Reaction Rate Ratio of 20, a Secondary Reaction Rate Ratio of 2 and A1/B1 of 1.0 ••••••••••••• 29 · 7. Data Generated with a Primary Reaction Rate Ratio of 20, a Secondary Reaction Rate Ratio of 2 and A1/B 1 of o.s ••••••••••••• 29 s. Data Generated with a Primary Reaction Rate Ratio of 2, a Secondary Reaction Rate Ratio . of 20 and A1/B 1 of 1.0 ••••••••••••• 29 9. Data Generated with a Primary Reaction Rate Ratio of 20, a Seoondary Reaction Rate Ratio of 2 and Ai/B1 of 1.0 ••••••••••••• 35 10. Data Generated with a Primary Reaction Rate Ratio of 20, a Secondary Reaction Rate Ratio of 2 and Ai /Bi of O • 5 • • • • • , • • • • • • • • 3 5 11. Data Generated with a Primary Reaction Rate Ratio of 20, a Secondary Reaction Rate Ratio of 2 and A1/B 1 of 2.0 ••••••••••••• 35 12. Products of the Conversion of Bromobenzene- sulfonic Acid to the Sulfonyl Chloride by Phosphorous Pentachloride •••••••• • • • 43

vi 'l'able Page 13. Products of the Conversion of Iodobenzene- sulfonic Acid to the sulfonyl Chloride by Phosphorus Pentachloride •••••••••••• 44 14. Experimental Data for the Relative Rate 46 Determination of' J.te~eft4\WBrmnobenzene. ' • • • • • 1.5. Relative Rates as Determined for Varying Initial Ratios of Benzene and Bromobenzene • • • 48 16. Equilibriu.~ Constant Calculation to Test Thermodynamic Control ••••••••••••• 50 17. Relative Rate Determination Utilizing Gas Chromatographic Separation of the sulfonyl Chlorides • • • • • • • • • • • • • • • • • • • 52 18. Relative Rate Determination Utilizing Labeled Benzene •••••••••• • • • • • • 53 Moles of the Final Products in the Competitive Sulfonation of Benzene and Bromobenzene •• • • 57 20. Counti.ng Data for the Isomer Distribution Determination • * ••••••••••• • • • • 62 21. Isomer Distribution of Bromobenzenesulfcnic Acid. 64 22. The Isomer Distributions in the Sulfonation of- the Halobenzenes • • • • • • • • • • • • • • 65 23. Ortho/Para Ratios for the Sulfonation of Bromobenzene •••••••••••••• • • • e 65

24. Ortho/Para Ratios for Runs 20 through 24. • • • • 66 25. Partial Rate Factors for the Sulfonation of Bromobenzene ••• • • • • • • • • • • • • • • • 67 26. Isomer Distribution and Partial Rate Factors for Low so3 Concentration •••••••• • • • 68 27. Precision of the Measurements Used in the Calculation of Relative Rates •••••• • • • 69 28. Melting Points of the s-Benzylisothiouronium Salts of the Bromobenzenesulfonic Acids •••• 74

vii -, /~ ,_ ' ;- ,,, A ., • ' LIST OF ILLtiSTRATIONS

Figure Page 1. Dewar's Proposed rr-Complex Mechanism for Electrophilic Aromatic Substitutions •• • • • • 2 2. Nelson and Brown•s Mechanism for Electrophilic Aromatic Substitutions ••••••••••• • • 6 3. Energy Diagram for a Typical Electrophilic Aromatic Substitution Reaction ••••• • • • • 7 4. Reactivity and Selectivity of Reagents on Toluene • • • • • • • • • • • • • • • • • • • • 13 s. Extension of the Reactivity-selectivity Diagram to Other Monosubstituted Benzenes ••••••• 13 6. Reactivity and selectivity of Electrophilic Reagents on Bromobenzene •••••••• • • • • 18 7. Christensen's Reaction Sequence for sulfur Trioxide sulfonation Reactions ••••• • • • • 26 a. secondary Reaction Effects in Mechanism I • • • • 30 9. Variations in ASH/BSH in Mechanism I •••• • • • 31 10. Secondary Reaction Effects in Mechanism II. • • • 36 11. Variations in ASH/BSH in Mechanism II • • • • • • 37 12. Chromatogram of the Products of the Brorno- benzenesulfonic Acid Conversion by Phosphorous Pentachloride ••••••••••••••••• 42 13. Apparent Relative Rates as a Function of Initial Benzene/Bromobenzene Ratios •••• e • 47 14. Variations in the Ratio of the Product sulfonic Acids with Sulfur Trioxide concentrations as Determined by Gas Chromatography •••••••• 54

viii Figure Page 15. Variations in the Ratio of the Product sulfonic Acids with Sulfur Trioxide Concentrations as 1 Determined by c 4 Analysis ••••••••• • • 55 16. Moles of Sulfonic Acid Products as a Function of Sulfur Trio..~id~ Cori~entration by Gas Chromatographic Anal~sis •••••••••• • • 58 17. Moles of sulfonic Acid Products as a FUnction of sulfur Trioxide Concentration by cl 4 Analysis •••••••••••••••••• 59 18. Isotope Exchange Reactor ••••• .- ••••••• 75 1 19. vacuum Line for c 4-Benzene Preparation • • • • • 76 20. The Sulfonation Apparatus •••••••••••• 78

21. Calibration of the Flowmeter • • • • • • • • • • • 106 22. The Bromination Apparatus • • • • • • • • • • • • 109

_i..x CHAPTER.I

INTRODUCTIONAND BACKGROUND

Electro~hilic Aromatic Substitution Considerable research has been dedicated to the detailing of the mechanism of electrophilic aromatic substitution reactions. Although significant success has been achieved in these efforts, there remain a number of points to be clarified. The observations that aromatic hydrocarbons exhibit a significant solubility in liquid hydrogen fluoride, 69 whereas aliphatic hydrocarbons are insoluble under the same conditions, led to the reasonable conclusion that an interaction between the HF and the aromatic system occurs. It was found subsequently that a number of systems showed similar reactions with the aromatic nucleus, silver ion,l hydrogen chloride and hydrogen bromide, 17 among others. McCaulay and Lien76 noted that the solubilit.y of a number of aromatic compounds in liquid HF was greatly increased upon the addition of boron trifluoride. Simj.larJ.y, it was observed that aluminum chloride was not by itself soluble in toluene, but upon the addition of HCl, the solution became intensely colorea. 16 Dissolving aluminun, bromide and HBr in various aromatic compounds gave similar

l 2 x·esults. The interactions of the mixed systems were much

stronger than with HF, HCl and HBr alone, as evidenced by the colo:r change., electl(;ig~l, cgpduotivity of the solution, 48 , ~t.. -,s,: :~-.g,. ,, . < and deuterium exchange which occurs in these but not in the ·systems with HX alone.71 The term, rr-complex, proposed by Dewar, 41 has been applied to the weaker interactions. He made the early

proposa1 4 2 that the mechanism of electrophilic aromatic substitution reactions involved, as its most important step, the formation of this rr-complex, I.

Fig. 1.--Dewar's proposed rr-complex mechanism for electrophilic aromatic substitution.

A single step mechanism wh_ich would not necessitate the disruption of the rr-electrai system of the aromatic ring was favored by Hamrnett5 7 since it was expected that considerable energy would be necessary to overcome the resonance stabilization of the aromatic ring system.

Dewar• s r.lechanism would require a primary kinetic isotope effect for all electrophilic aromatic substitution reactions. While a small secondary isotope effect has been 3 noted in some cases 38 •73 , 91 very few reactions studied to this point show a significant primary effect, 60 and in most, no effect is noted. 7 ,SO,Sl The question of diffusion control

in instances when a small isotope; effect was noted in reactions·with benzene and hexadeuterobenzene was considered by Bosscher and Cerfontain, 9 who studied the sulfonation of 1,3,5-trideuterobenzene and analyzed for the competit:i.on between protium and deuterium on the same molecule. A primary isotope was not noted. Since a primary kinetic isotope effect is not characteristic of all electrophilic aromatic substitution reactfons, the obvious conclusion is that the rate limiting step must occur later than the initial interaction of the electrophile and the aromatic rr-electron system. The $tronger complexes, as described above, are thought to involve the association of the electrophile with a particular carbon atom in the aromatic ring. This, of course, disrupts the aromatic electron system, forming a cyclohexadienyl cation, termed a cr -complex, II. So far as the energy requirement for ·overcoming the resonance

II 4 energy of the aromatic system is concerned, there is to be expected some residual resonance energy in the cyclohexadienyl cation. By the Huckel treatment of rr-electron systems, 75 values of 2 S for the resonance energy of benzene and 1.46 ~ for the eyclohexadienyl cation are obtained. The difference between these two energies is not unreasonable for the kinetically significant formation of the cr-complex. Although a partial correlatio..~ is found with the stabilities of the rr-complexes, the data in Table l show a better correlation between halogenation rates and the cr -comple.."< stabilities.

Table 1.--correlation of Halogenation Rates with Complex Stabilitiesa

Halogenation Benzene TT-complexb cr -complexc Rated

Unsubstituted 0.61 o.o9 0.0004 Methyl' 0.92 0.63 0.24 1, 4-•Dimethyl 1.00 1.00 1.00 1,2-Dirnethyl 1.13 1.1 2.1 1,3-Dimethyl 1.26 26 204 1,2,4-Trimethyl 1.36 63 ~00 1,2,3-Trimethyl 1.46 69 660 1,2,4,S-Tetramethyl 140 1,100 1,2,3,4-Tetramethyl 1.63 400 4,400 1,3,5-Trimethyl 1.59 13,000 75,000 1,2,3,5-Tetramethyl 1.67 16,000 170,000 Penta.methyl 29,000 360,000

aRelative top-Xylene. baeference 17. CJleference 67. dReference 23. 5 Perhaps the most significant evidence supporting the role of the cr -complexes has been the isolation of some particularly stable species. The salt, III, was obtain.ed from a mixture of benzotrifluoride, nitryl fluoride and boron trifluoride and decompoa,a upon heating to give the nitrated benzotrifluoride.97 This may be considered to be representative of similar, though leas stable intermediates in other aromatic substitution reactions •

. CF3

III

Nelson and Brown86 have proposed a generally · comprehensive mechanism which allows for the variations and peculiarities noted in the broad spectrum of electrophilic aromatic substitution reactions (Figure 2). The initial interaction of the electrophile and the aromatic TT-electrons results in a rapid equilibrium

formation. of a TT-complex (1). This TT -complex is represented by a shallow minimum, A, on the energy profile shown in Figure 3. As the reaction continues along the reaction coordinate, a cr -complex is formed (2), which is a reactive intermediate, a relatively deep minimum, B, in the energy diagram. The fate of the a-complex may be a 6

) (2) (. I ~z+ CJ-z..,,I

z < > (3)

( > (4)

Fig. 2.--Nelson and Brown's mechanism of el.ectro- philic aromatic substitution. return to the previous rr-complex or the formation. of a new rr-complex (3), involving the departing hydrogen and the aromatic rr-system. The hydrogen may then be removed from the complex to canplete the reaction by the solvent or the base originally associated with the incaning electrophile

(4). Other energy diagrams have been proposed based on slightly different energy relationships. Olah 94 proposed a high activation energy required for the formation of the rr-complex for reactions which show an isotope effect. 7

Fig. 3. -Ene.rgy diagram for a typical electrophi.lic aromatic substitution reaction.

Caille ·and Corriu, 25 restudying brominations and chlorinations of benzene and toluene, took ~.xception.to Olah's proposal of a rate-determining rr-complex formation in these reactions and suggested that the rate of reaction is determined by the formation of a " cr -compl~ activated state." On the basis of increasing isotope effect with increasing steric hindrance. Nilsson 91 proposed (5) as a mechani-sm for the bromination of various l, 5-disubstituted benzenes with k2 being slower or of about the same rate as 8

~+ k2 ArH + Br2 ~ ArHBr + Br- --ArBr + HBr (5)

Reactivity and Selectivity Early in the studies of the aromatic system, it was found that substituents on the ring caused a predominance of substitution in the ortho and para positions, or in the meta position depending upon the nature of the substituent. The substi.tuent also affected the overall reactivity of the aromatic system. Inductive and resonance interactions between the ring and some substituents are believed to increase the electron density at the ortho and para positions leading to the observation of ring activation and ortho and para direction. Some substituents are found to be ring deactivating but ortho and para directing due to a combination of overall inductive withdrawal of electrons from the ring and resonance stabilization at the ortho and para positions. Other substituents combine depletion of the electron density at the ortho and para positions with overall inductive withdrawal to give a deactivated aromatic system on which the predominant electrophilic substitution occurs in the meta position. Several "rules of thumb" were 92 proposed to predict these substituent effects. 2 • A particularly dramatic demonstration of these effects was the high percentage of m-cymene produced in the Friedel-crafts isopropylation of toluene. 39 These results 9 prompted the formulation and proposition of the reactivity- selectivity principle by Bro"1n and Nelson.20 Briefly stated, the reactivity-selectivity principle proposes that the more reactive an attacking electrophile is, the less selective it will be as to the choice of ar~~~tic substrate·and also as to the position of a given aromatic molecule. The intermediate in the Friedel-crafts isopropylation reaction is considered to be a very reactive electrophile 87 , 88 and, hence, would be little influenced by the slight variations in electron density about the various positions in a given aromatic system, or, for that matter, between various kinds of aromatic molecules. on the other hand, a bromine moleo.1le is considered to be weakly electrophiU.c so that even slight variations in electron density will have significant influence on the rate and position of its substitution. The data in Table 2 show the decrease in selectivity both as to the position of substitution in toluene and also between toluene and benzene as the reactivity of the electrophile increases. To elaborate the details of reaction rates due to the substituent effects and electrophile reactivities, it is helpful to compare the partial rate factors for the various positions. These give the differences in reaction rates for each for the various positions of a monosubstituted benzene as compared to one position in unsubstituted benzene (6). 10 Table 2.--selectivity and Reactivity of Electrophilic Substitutions on Toluenea

Reaction kTkB % Meta

Bromination Br2 in HOAc,at 25° 467 0 Chlorination Cl2 in HOAc at 250 353 o.s Chloromethylation CH20, ZnCl2, HCl in HOAC at 6QO. 112 1.3 Nitration AcON02 in Ac2 at o0 27 3.7 0 Methylation MeBr, JUBr 3 at o 3.5b 17.3 Isopropylation C3Hs, A1Br3 in CH3N02 2.1 29.ac asummarized in Reference 86. bAt 40°. CAt 0°.

% p-isomer ~onosubstituted benzene 6 X (6) Pf= s X 20% kbenzene

The six positions of benzene are compared to the three statistically weighted distinguishable (two ortho, two meta and one para) positions in the monosubstituted benzene. The percent p-isomer actually obtained divided by the statistical 20% (one p-position of the five positions available) indicates the proportional preference of the electrophile for that particular position. The reaction rate constant ratio of the substitution of the monosubstituted benzene and benzene, combined with the other factors, gives the rate of para substitution relative to the rate of the same substitution in one position of benzene. Ortho- and meta-factors are obtained in the same .. mannt~r, considering the two equivalent positj_ons for each. 11

% o-isomer ~onosubstituted benzene Of= ~ X X (7) 40% kbenz.e~e

% m-isomer kxnonosubstituted benzene - 6 X X (8) - '5 40% 'kbenzene

pf A linear relationship between log Pf and log~• which may also be derived77 from Hammett type equations (9-13), was found by Brown and Nelson. 20

+ log Pf = fCJp (9)

log mf = p CJm + (10)

Pf + + log- = ( CJp CJm ) (11) ffif f -

From these equations, then

l Pf log Pf ( (12) log-+ = mf aP (J p + - crm+> or p+ log Pf = log Pf ( =J=) (13) mf CJp + - O'm

This is the equation of a straight line passing through the origin with a slope of (- o E?+ ) • The crp+ - am+ validity of this treatment was tested in its application to toluene as given in Figure 4. Since then, other systems have been studied to test the fit to a line given by 12

0 + (- P ) for the various monosubstituted benzenes, (the a p+ - am+ values have been tabulated by Brown and Okamoto 22 ). several projections for other systems·are shown in Figure 5.

Log Ptlmt may be ~houg~t of as a measure of the selectivity of the incoming electrophile while the log Pf is considered·as a measure of the activity of t~e reagent. Although the reactivity-selectivity principle has been successfully applied to a large number of systems, there have been some significant deviations found. 93 It has been suggested that the reactiv,ity-selectivity principle holds pnly in reactions where the formation of the u -can- plex is rate determining and will not apply·in other situations. 95 , 104

Electroohilic Aromatic Substitution Reactions on Brornobenzene Although not as completely studied as toluene, a considerable amount of research on the electrophilic substitutions on bromobenzene has been reported. Table 3 presents the data for a number of substitutions which'may be utilized in a reactivity-selectivity plot for brornobenzene.

This plot is given in Figure 6 and includes the line giver1 by equation .(13) where crp+ = 0.150 and cr + + 0 • 40 5 as m reported by Brown and Okamoto 22 for the substituent effects of the bromo substituent. The slope of the line is -0.59. As can be seen, a number of points approach this line; however, a large degree of scatter is very obvious. 13 / l ochloromethylation. I . 0 basic·i· ty (BF3-HF) L 1ttlt1otl. oP} mercuration 0 /detrirnethyl silylation / .· O sulfonylation isopropylation

log Pf/Inf

Fig. 4.-Reactivity and selectivity of reagents on toluene (from Reference 88). ;

a t-Butylbenzene , b Toluene c Ethyl phenylacetate d Benzyl chloride e Chlorobenzene f Benzal chloride g Benzotrichloride h Ethyl benzoate i Nitrobenzene

0 log Pf/l'Uf

Fig. s.--Extension of the reactivity-selectivity diagram t.o other rnonosubstituted benzenes (from Reference 88). Table 3.--Partial Rate Factors for Electrophilic Substitutions on Bromobenzene

Conditions log Pf log ft Ref, No. Reaction Of fflf Pf fflf

- 0 l Acetylation CH3COCl, C2H4Cl2, AlCl3, 2s c 0.84 -.074 19 0 2 Benzylation ¢CH2Cl, MeN02, AlCl3, 2s c 0.175 0.00378 0.721 -.142 2.280 101

3 Bromination Br2, MeN02, HOAc, 25°c 0.00053 0.0610 -1.208 1.214 63 0 4 Bromination ar 2 (neat), MeNo2 , Fec1 3 , 2s c 0.212 0.0001a 0.13a -.860 2.885 102 5 Bromination Br2 (in MeN02), FeCl3, 2s0 c 0.010 0.00006 O.444 -.3-~ 3.868 102 0 6 Bromination Br2 & FeCl3(in MeN02),MeN02,is c 0.1306 0 .•00096 0.699 --lt?-6 2.844 102 o· O.259 .2.936 49 7 Bromination Br2, CS2, AlBr3, 54.7 C 0.0402 0.0003 -.5&7,";" "

8 Bromodeboro- nation ArB(OH)2, Br2, HOAc (20%), 2s 0 c 0.044 0.413 --~ .973 74 9 Bromodesily- lation ArSiMe3, Br2, HOAc (98.5%),2s 0 c 0.011 -1.149 47 10 1-Butylation i-BuBr, MeN02, SnCl4, 2s0 c 0.12 -.921 96 11 1-Butylation ~-BuBr, MeN02, AlCl3, 2s0 c 0.0018 0.116 -.936 1.809 96 12 1-Butylation i-Butylene, MeN02, AlCl3, 25°c 0.18 -.745 96 13 Chlorination Cl2, MeN02,'FeCl3, 2s0 c 0.197 0.0144 O.47O -.328 1.513 103 0 0.196 0.0135 0.480 -.319 1.564 103 14 Chlorination Cl2, MeN02, A1Cl3, 2s c ..., ~ Table 3 (continued)

No. Reaction Conditions log p..: log~ Ref Of Inf Pf ,L. mf •

15 Chlorination Cl2, HOAc (aq.), 2s0 c 0.0838 0.0032 0.310 -.509 1.986 111 0 16 Destannylatioi 1 ArSnMe3, EtOH•H20, HCl04, so c o.145 -.839 45 17 Ethylation EtBr, C2H4Cl2, GaBr3, 25°c o.oa1 o.433 -.363 .698 21 0 18 Mercuration Hg(OAc)2, HOAc, 2s c o.os4 0.21 -.569 .699 18 19 Mercuration Hg(OCOCF3)2, CF3COOH, 2s0 c 0.0126 0.00954 0.194 -.612 1.308 24 0 0.00098 0.103 -.988 2.012 6 20 Nitration HNo3 , MeNo2 2s c --". 21 Nitration .. N02+ BF4, - (CH2)4S02, 25°C 0.0925 0.00396 0.264 -.579 l.8·i3 100 '. ., 22 Nitration N02+ PF6 - , MeN02, 25 0 C ..__ 0.399 0.772 -.112 98 . ·. 23 Nitration CH3COON02, (Ac)20, 1s0 c 0.033 0.0011 0.112 -.951 2.0«6 6

24 Protodesily- . . 0 lation ArSiMe3, MeOH•H20, HCl04,Sl.2 C ' 0.10 -1.0 44

25 Protoboro- 0 nation ArB(OH)2, H20-H2S04, 25 C 0.30 -.523 85 0 26, 1-Propylation i-Propylbromide,MeN0 2 ,AlC13,2S c 0.123 0.211 0.179 -.748 .819 96 27 i-Propylation i-Propylbromide,MeN02,FeCl3,25°c 0.171 0.0383 0.2416 -.616 .soo 96 0 28 i-Propyla:t:.ion Propylene, MeN02, AlCl3, 2s c 0.116 0.0181 0.1512 ~.821 .,,. .921. 96 0 29 sulfonation S03, S02, 12 C this work

''"• ..., U'I 16 A considerable amount of scatter was observed in similar electrophilic aromatic substitutions on chlorobenzene.lO In some cases, thoug~ ~ot all, those largely deviant. bromobenzene reactions correlate with similarly deviant chlorobenzene reactions. Table 4 lists several reactions having the widest deviations (in decreasing order) from the line for bromobenzene and from the corresponding line (slope= -0.4) for chlorobenzene under the same reaction conditions.

Table 4.--Reactions Having Large Deviations from the Reactivity-selectivity Relationship

Bromobenzene' Chlorobenzene Reaction No. Reaction No.b

la 5 7

2 6 22

3 2 21

4 7 5

5 4 12

6 14 3

7 13 4

8 12 11

9 21 19 lO 3 10

aLargest deviation and in decreasing order. bChloroLenzene reaction no. corresponds to the same reaction conditions as the corresponding bromobenzene reaction no. and is usually reported in the same source. 17 seven of the ten most deviant results for the bromobenzene series of reactions are also among the ._ten most deviating of the chlorobenzene series. Undoubte,Uy, one significant factor in the large amount of scatter is the Slftal.i?ililtoufit of meta-isomer formed ·, in these electrophilic reactions with bromobenzene and chlorobenzene. In some reactions the percent of meta product is reported as " { 0.2 percent." Obviously, experimental error will have a great influence on the result. The difference between 0.2 and o.4- percent is, in most cases, less than e:cperimental error, yet causes var:tations of 100% in the para/meta ratio. Those reactions giving results nearest the indicated line are 17, 18, 27, 11, 19, 23, 26, a_nd 28. These reactions are alkylations, mei:;curations, and conventional nitrations which are typically fairly reactive. This was found to·be generally true in the reactions of chlorobenzene. 02 06

~3 os

'M o. 5 Ql4 ~· ' 'I"~. 021 g' o. 6 o' .... . ()19 0.1 2~c;o Q25 ~ o.a 027 ~ c4 0.9~ '\. 0ll~a 1.0 L '\. 20

1.1 ·. 1. 2 o3 o.oL 0.2, 0.4. 0~6, OJ3 , l•Q, ·l.2 lA J.6 l.8 .2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 log Pflmf ... Fig. 6.--Reactivity and selectivity of electrophilic reagents on bromobenzene. (X) CHAPTERII

AROMATICSULFONATION

The sulfonation of aromatic compounds has been of significant economic interest for the past centu~y in the refining of crude petroleum fractions. More recently, it has become increasingly important as a method of purification of aromatic hydrocarbons, in the preparation of a

,, variety of _intermediates, in addition to increased interest in sulfur chemistry in generai ..50

Nearly all sulfonating agents are so3-derived. sulfuric acid, the S03 hydrate, in various concentrations, and oleum., so 3 dissolved in 100% sulfuric acid, are the most generally used industrial sulfonation agents. Chloro- sulfonic acid, which in many respects may be more accurately described as s03•Hc1, 51 is often used in laboratory scale sulfonations. sulfur trioxide, itself, has only relatively recently become readily avai~able and is one of the more reactive sulfonating reagents. There have bee~ no less than forty-five s03-organic compound complexes used in sulfonation reactions.5 1 Sulfonations in concentrated sulfuric ac'id appear to involve an S03 molecule as the reacting species since the rates of reaction are inversely proportional to the square 20 of the concentration of water added to the system. 40

(14)

(15)

(S03) (H30+) (HS04-) l

2 These equations show the l/(H 2o) relationship noted experimentally, but sj.nce the assumptions that the a2so4 is unionized and that the activity coefficients are unchanged· with the increasing water added, are very probabiy invalid, the evidence for so3 as the sulfonat.ing species is not u..~equivoca1.S 6 Gold and Satchel, 55 after having considered these objections in a complex, treatment, sugges·t, though ..· reservedly, that the aulfonating agent is likely so3 • Sulfonaticns, especially when so3 is used as the sulfonating agent, are somewhat atypical as an electrophilic aromatic substitution reaction. Among the unique characteristics of these reactions are their reversibility under relatively mild conditions, a demonstrated kinetic isotope effect in most cases, a kinetic order of 2 for the sulfur trioxide and the complication by possibly significant secondary reactions. 21 J!~,;ersibili tz It has been suggested89 that the property of the int.ern-.ed1.ate which le&~ii~\;; .. ~ility may also underl,ie ~he other unusual chara.cteristics ·of su.lfonation reactions.

Since the so3 molecule is 11eutral, the attack on the neut.ral ci.roir«.,tic sys-c.em results in an electrically neutral o ••complex, IV.. This neut~ality possibly ·1mparts a st.i:::r.,il:.H:.y to th,:1 a-complex that results in a reduce:d leaving tenderir;y for the hydrogen ion (l 7).

so - .,....---~ (1,7) -+ ___. a+ S03 + ,;.-

Holleman and Caland 62 fowid that at 100°c in su.lfurlc acidt m-toluenesulfonic acid is unchanged but that. 2- ar:d p'!"#toluenesu.lfonlc a.cids are interconverted. A recer1t study 117 of the revi'.!rsibi.lity of the sulfonation of t-oluene revealed that an equilibrium concentration of abou.t 45% para, 5% ortho and 50% meta isomers results when aqueous sulfuric ac:f.d solutions of the toluenesulfonic acids are heated to 14o0 c. · These same workers found that the first order isomerization rate cor1stants compared favorably with the desulfonation rate constants. 116 22 The kinetics of desulfonation of benzene and toluene 116 were reported with partial rate factors at 120° and l40°c. Table 5 shows t~ese. data along with the. isomerization data.

Table 5.--Isomerization and Desulfonation of Toluenesulfonic Acids

Pf Of Mf Ptlmf. log Pf/mf

Isomerization H2S04 (aq.) 140°c 45 1.635 Proto- desulfonation H2s04 (aq.) 12b 0 c 60 154 2.2 27.3 1.435 Proto-. desulfonation H2S04 (aq.) 140°c 80 110 2.9 '27.6 1.414

The desulfonation of other arylsulfonic acids 3 82 3 110 66 82 106 including Y = C2Hs~ OC2H5, NH2, ' c1, OH, COOH,

4 62 S03H, and ca3 has been studied.

Kinetic Isotope Effects Although an isotope effect was observed in the nitration of 2,4,6-trisubstit~ted benzenes, 84 this is the exception rather than the rule. Brominations, nitrations and most other electrophilic aromatic substitu~ion reactions do not exhibit a kinetic isotope effect. 54 23 Sulfonation reactions usually show a small but significant isotope effect.S,Sl ;,

Al though a maxim~: ;pri~~"-~8/k 0 1 s estimated at 6. 9 at 20°c, 58 s_ome relati.vely slow reactions have exhibited small primary isotope effects where the effect is thought to be due·to an unsymmetrical transition state--the incaning electrophile probably is held'more strongly than the hydrogen (or deuterium). These small primary isotope effects are not kinetically distinguishable from secondary isotope effects. If k8/k 0 is greater ihan about 2.0, one may be fairly certain that the isot.ope effect is a primary one, but beyond this, it would be difficult to tell whether or not there is a hydrogen transfer in the rate controlling•step. 60 Secondary isotope effects are t~ought to arise from. hyper- conjugation and inductive effects and the variations in the balance of hybridization in the transition state.70 A kinetic isotope effect in aromatic sulfonation was first reported in 1949 when Melander observed that in tritiwn labeled benzene and bromobenzene, the tritium atoms were replaced much more slowly than the protium atoms. 79 Recent studies have provided more information on these effects. Kort and Cerfontain7 3 have observed variations of

An k8/k 0 with varying concentrations of sulfuric -acid. observed ratio as high as 2.5 indicates that the kinetic isotope effect is probably a primary one. In an earlier experiment, they reported that in the sulfonation of a mixture of benzene and hexadeuterobenzene, no primary 24 isotope effect was notea, 28 and concluded that where the pyrosulfon:ic acid is an intermediate in the sulfonation process, the kinetic isotope effect usually observed is a seco~dary one, originating in the steps leading to_ the intermediate v.

Secondar~ Reactions It has been recognized for sane time that secondary reactions do occur in the sulfonation of aromatic compounds with sulfur trioxide. These are often completely ignored or inadequately considered. At best, their function and influence are not clearly understood. Under most reaction conditions, three products are isolable: 51 the monosulfonic acid (disulfonation is not normally a complicating factor}, diaryl sulfones, which are very minor products under the usual reaction conditions, 8 ,llJ0 9 and sulfuric acid. Besides these, pyrosulfonic acids are postulated as being intermediate species, but, because of their extreme reactivity, ar~ not isolable. These pyrosulfonic acids may be the products directly derived from the·· r~action Hinshelwood43.11 4 ,llS observed as ~eing second 25 order in sulfur trioxide. Sulfur trioxide is ttsually monomeric in solution5 3 but the second molecule does not appear to function as a typical base in extracting the proton. 61 Cerfontain 8 , 9 , 3o concluded that the formation of the pyrosulfonic acid 'Was a ra,P:i.d initial step which used up half the aromatic compound (originally equimolar with·S03) and that a slower step subsequently produced the arylsulfonic acid. Hinshelwood reported a "retardation by the product" after the initial stages of the reaction. Christensen 35 observed the formation of a colored 111 iodobenzenesulfonic acid-sulfur trioxide species which he found to be very react:.tve as a sulfonation agent. Pyrosulfonic acids appear to be the intermediates in the formatie>n of the sulfones, 30 , 34 , 109 which must be, then,, slightly competitive with the sulfonation by the pyro- '· . sulfonic acid (18). Because of this subsequent sulfonation,

ArH + 2 S03

the pyrosulfonic acids are often ignored in kinetic investigations, being considered as reactive inter.mediates in the overall reaction. Christensen has also studied the formation and reactivity of sulfonic acid anhydrides. He concludes-that the anhydrides are somewhat less reactive as sulfonation 26 34 reagents than the pyrostilfonic acids and s0 3 , but he proposes that much of the sulfonic acid produced in the end has probably passed through the anhydride stage. While sulfonic acid.anhydrides are stable enough to be isolable, they are very· t'eabt1.fte~2 and are better derivatization agents than the sulfonyl chlorides. 37 Although the concentrations o~ anhydrides are unusually low and considered insignificant at short reaction times, 9 Christensen suggests that they are actually very significant but that their reactivity makes it easy to overlook them. 37 He proposes the following sequence for the overall system. 36

RH+ 2

Fig. 7.--Christensen•s reaction sequence for sulfur trioxide sulfonation reactions.

Secondary Reactions and Their Effects on the Final Product Distribution Unless the sulfonating species involved in the secondary reaction has precisely the same reactivity (and selectivity) as the primary sulfonating agen~, ~he final distribution of products will ·be different than in the absence of the secondary reaction. When differences occur, 27 the effects of the secondary reaction would be expected to be as follows, if the secondary sulfonating agent is less

:·· ...:." , , ..~t:·t"'· -,:, .:11· ►.-,:r· , reactive, i.e., more se!ective", th~h the primary reagent, the final product distribution will show a greater quantity of the product derived from the more reactive substrate; if the secondary reaction is more reactive (less selective) than the primary reaction, then the effect of the secondary reaction will be to increase proportionally the product formed in the smallest amount in the primary reaction •. To show the effects of'secondary reactions, data were artificially generated using rate constant ratios of k,AfkB = 20 and k_A/kB = 2 and calculating the product and product ratios for each of the above situations. According to the most recent consensus, the primary reaction probably results in the pyrosulfonic acid which ~ . . then undergoes a secondary reaction, the sulfonation of. more of the starting aromatic compounds, however at a slower, i.e., "retarded," rate. A fu1:ther assumption is necessary as to the relative reactivity of the products'of the primary reaction in the secondary sulfonation reactions. When the rate constant ratio is large, the primary product of the less reactive aromatic will be small so that only small error is introduced by assuming equal reactivity Qf the primary products. If it is also assumed that the slower secondary reaction uses up half of the total pri..~ary products as they are produced, a "typical" final product distribution may be calculated. _Details of the calculation 28 are discussed in the Appendix.

The primary reactions of Mechanism I, then, may be represented as

AH + 2 s ASSH (19) BH + 2 s BSSH (20)

<'I and the secondary reactions as

ASSH + AH 2 ASH (21) ASSH + BH ASH + BSH (22) . BSSH + AH ASH + BSH (23) BSSH + BH 2 BSH (24)

From these conditions, the data shown in Tables 6, 7 and 8 were obtained and are shown graphically in Figure 8, which shows the amounts of the,products, and Figure 9, which shows the ratios of the products. The departure from the straight lines'· given in each figure indicates the contribution and effects of the secondary reactions •. The plot of the product ratios with increasing extent of reaction does show the increasing production of ·the final product of the less reactive aromatic. It is expected and is obvious from the data that the influence of the secondary reactions is greater towards the end of the reaction or as the concentration of the more reactive substrate is depleted. This same effect is demonstrated in a larger ratio of products from t.~e more 29 Table 6.--Data Generated with a Primary Reaction Rate Ratio of 20, a .Secondary Reaction Rate Ratio of 2 and A1/B1 of l.O

ASH Af Bf, ASSH BSSH ASH BSH ASH+ ·• ', ;,. .,,,.:: BSH BSH 17.55 19.73 1.so o.oa 0.95 0.19 l·.14 s.oo 14.67 19.17 2.25 0.13. 3.08 0.10 3.78 4.40 11.52 18.46 2.63 0.17 5.85 l.37 7.22 4.27 s.22. 17.63 2.81 0.22 8.97 2.15 11.'i2 4.17 4.83 16.69 2.90 0.30 12.27 3.01 15.28 4.08 2.56 15.56 2.20 0.30 15.24 3.84 19.08 3.97 .,

Table 7.--Data Generated with a Primary Reaction Rate Ratio of 20, a s.econdary Reaction Rate Ratio of 2 and A1/B1 of o.s

Af Bf ASSH BSSH ASH BSH ASH+ ASH BSH -BSH 17.63 39.62 1.so 0.15. 0,87 0.23 1.10 3.78 14.71 38.92 2.27 0.24 3.02 0.84 3.86 3.60 11.50 38.03 2.64 0.32 5.86 1.64 7.50 3.57 8.11 36.96 2.82 0.44 9.07 2.59 11.66 ·3.,50 4.60 35.69 2.91 0.61 12.49 3.69 16.18 3.38 2.19 34.55 2.21 0.60 15.60 4.82 20.42 3.24

Table 8.--oata Generated with a Primary Reaction Rate Ratio of 2, a secondary Reaction Rate Ratio of 20 and Ai/Bi of 1.0 ,. ASH Bf ASSH BSSH ASH BSH ASH+ Af BSH FsH

17.28 · 18.94 . 1.49 0.11 1.23 0.29 1.52 4.24 13.46 17.71 2.25 1.24 4.29 1.os 5.34 ,4.09 8.99 16.21 2.62 1.66 8.39 2.13 10.52 3.,94 4.02 · 14.15 2.81 2.26 13.17 3.59 16.76 3.67 30

19 0 _.,_A1/B1= 1.0 18 ct A1/B1 :: 0.5 .. N 17 0 A1/B1 = 1.0

:i:: 10 ~ 'O 9 ij :i:: ·8 U) IQ 7

0 1 2 4 6 a to t2 !4 !'6 1a 20 22 2 Extent cf reaction

\ Fig. s.--secondary reaction effects in Mechanism I. 12 ~ A1/B1 • 0.5 kAl/kal = 20,: kA_/k;a :'a 2 ,. 2 2 11 ··, kp,. ✓kB 0 A1/B1 = l. 0 kA1/kBl = 20, 2 = 2 A1/B1 l.O kA /ks 2, 10 CJ = 1 1 = kA2/kB2 = 20

Cl) Cl). 6 =r:a< Ill 5 0 4 0 0 Q'--··~-,01------=---~----0- 0-::-:::::==="---~" □- ----__ o----=---o" □---- Q 3

0 l · 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Extent of Reaction

Fig. 9.--variations in ASH/BSH in Mechanism I. ...,w 32 reactive relative to the le~s reactive substrate at high initial ratios, and also at 1ow initial concentrations of the electrophile. The influence of the secondary reactions upon the product distribution is less significant when the secondary' reaction is more selective (kifkB = 20) than the primary reaction (kJ/kB = 2) than when the secondary reaction is less selective. The slight influence of the secondary reaction under these conditions is less unexpected than the apparent (slight) decrease in the ASH/BSH.ratio. As an alternative to the previously discussed mechanism I, it may be postulated that the products of the primary reacti.ons are, in f;a.ct, final products, some of which become involved in secondary reactions. 1'he products, .oft.he secondary reaction may react with more of the starting reagents (with a different selectivity than in the primary reaction) to give final products in a different ratio than obtained from the primary reaction. Such a mechanism (II) may be outlined as follows: the primary reactions are

AH + s ASH (25) BH + s BSH (26)

A portion of these products may then react with additional electrophile to give as secondary products ASH + s ASSH (27) BSH + s -----+ BSSH (28) 33 The se:condary products may finally react with more of·the aromatic reactants to give final products as obtained in the primary reaction ,,::.,:;

ASSH + AH 2 ASH (29) ASSH + BH ASH + BSH (30)

BSSH + AH ASH + BSH (31) BSSH + BH 2 BSH (32)

Those final products derived through the secondary reactions may return again to the secondary reaction cycle. In order to generate data to test this mecha.,nism it ,, is necessary to define various conditions and make certain assumptions. These area (l) the reaction rate ratio for the primary reaction and secondary reactions, (2) the relative portion of the primary products which become involved in the secondary reactions (these must include a "recycling" of the final products of the secondary reactions), (3) the relative reactivity of the prinrary products in the secondary reactions, (4) the reactivity of the secondary products with the starting aromatics, and (5) the selectivity of the final reactions. In the generation of the data in Tables 9, 10 and 11, the relative rate ratio for the primary reaction was chosen as twenty. one-fourth of the tota.l of the primary products were assumed to be involved in the secondary reactions whlch were assumed to be non-discriminatory between ASH and BSH. The reaction between the secondary products and starting 34 aromatics was considered to be complete--the concentration of the secondary products remained nearly zero--and the reaction rate ratio fo:t the r~iicH:.:l6bof the secondary products with the starting aromatics was chosen as two. ,, Details of the mechanics of the data generation are given in the ~pendix. 'l'he data presented· in Tables 9, 10 and 11 are plotted in Figures 10 and 11. It is readily apparent that these data more nearly approximate the behaviour of the experimental data as presented in the Results and Discussion

') section. The fact that the data obtained from ~~is proposed mechanism give plots distinctively similar to the experimental data, whereas the data obtained fran the mechanism suggested by the literature and considered earlj,er are only remotely similar, indicates that further consideration is necessary.

Secondary Reactions and Their Effects on Isomer Distribution The'influence of the se<;ondary reactions on the isomer distribution of toluene- and chlorobenzenesulfonic acids has been considerea.14 Data reported by Cerfontain 26 , 27 , 29 show a decreasing ortho/para ratio with increasing conversion of toluene to toluenesulfonic acid. The conditions of the reaction involved bubbling sulfur trioxide through neat toluene. With a constant concentration of SO3 thus supplied, the pyrosulfonic acid and 35 Table· 9.--oata Generated with a Primary Reaction Rate Ratio of 20, a Secondary Reaction Rate Ratio of 2 an_d A1/B1 of 1. O

Af Bf ASH BSH ASH+ ASH BSH BSH 17.83 19~81 2.1,. 0.1, 2.36 11.42 15.29 19.42 4.74 ci.58 S.29 8.12 12.02 18.62 8.98 1.38 10.36 6.50 s.11 17.50 12.89 . 2.50 15.39 5.15 3.36 15.88 17.64 4.12 ~l.76 4.28

Table 10.--oata Generated with a Primary Reaction Rate Ratio of 20, a Secondary Reaction Rate Ratio of 2 and A1/B 1 of 0.5

•. ASH Af Bf ASH· BSH ASH+ BSH BsH

17.82 '9.86 2.18 O.14 2.23 15.57 15.26 9.52 4.74 0.48 5 ..22 9.88 12.28 8.97 7.78 1.07 a.as 7.27 8.63 8.07 11.43 1.97 13.40 s.0O 4.22 6.75 15.84 3.29 19.13 4.81

Table 11.--Data Generated with a Primary Reaction Rate Ratio of 20, a secondary Reaction Rate Ratio of 2 and A1/B1 of 2.0

ASH BSH ASH+ ASH Af Bf BSH BSii 17.81 39.71 2.19 0.29 2.48 7.55 15.22 39.18 4.78 0.82 5.60 s.s4 12.11 38.35 7.89 1.65 9.54 4.78 8.34 37.11 11.66 2.89 14.55 4.03 3.71 35.27 16.29 4.73 21.02 3.44 36

19 ' C,A1/B1 = 2.0 18 0 A1/B1 1.0 = 0 ~ o.s 17 A1/B1 = ct 16 '- 15

13 0 12 Q C, 11

:.:: 10 ell < ttf 9 0 ~ :i: 8 Cll ~ 7

5 '· () 4 0

~ 3 C, 0 2 Q

0 4 6 8 10 12 14 16 18 20 22 24 26 EXtent of reaction

Fig. 10.--secondary reaction effects on Mechanism II. •. 13

11 (i A1/B1 • 2.0 10 0 A1/B1 = 1.,0 () A1/B1 = 0.5 9

7 :i:rCl'Hll <~ 6

5 Q- 4,

3 --

0 1 2 3 4 5 . 6 7 8 9 10 .·ll 12 13 14 15 16 17 18 19 20 Extent of Reaction w Fig. 11.--variations in ASH/BSH in Mechanism II. -.J 38 and sulfonic acid anhydride would be expected to 1ncrease36 as the reaction proceeded.

(33)

(34)

If, then, the change in the isomer .distribution is due to secondary reactions, the result would be a decrease in the ortho/para ratio in the latter stages of the reaction where the secondary reactions become prominent. This effect, the decreasing ortho/para ratio, is easily rationalized through the greater steric interactions of the physically larger secondary sulfonating agents. ., \ CHAPTERIII

RESULTS AND DISCUSSION

The sulfonation of aromatic compounds by sulfur trioxide in liquid sulfur dioxide is a straight-forward and relatively uncomplicated reaction from an experimental standpoint. Under the experimental conditions utilized in this study,·a high hydrocarbon/sulfur trioxide ratio, relatively dilute concentrations, and low temperatures, 109 limit the formation of sulfones. The sulfones are removed during the work-up after the evaporation of the sulfur dioxide solvent during the extraction (from ether by water) of the other products of the sulfonation. Although the sulfones are removed, the sulfuric acid produced along with the sulfones3 6, 109 is titrated and, hence, accounts for the

(35) molecules of aranatic compound involved. This net effect will not introduce any error in the total moles of aromatic compound used up in the reaction, but will. give a low yield of the grams of final product by a factor of 96/218 for the diphenyl sulfone, 96/476 for the dibromodiphenyl sulfone and

39 40 96/397 for the mixed sulfaie produced in the reaction. The ·significance of these considerations is limited by the very__ low yield.of sulfone ih ttt•;refftloft, Since sulfonic acid .anhydrides are very readily hydrolyzed3 6 to the sulfonic acids, it is expected thc;tt any produced in the reaction and,not used up as a secondary sulfonating reagent would appear at the titration step as the sulfonic,acid. If the production of the anhydride is by the mechanism (36) suggested by Christensen,36,there will be sane slight error introduced. For each anhydride finally

(36) titrated, assuming the so3 produced at the same time goes on to further sulfonation, there will be fi,,e protons titrated for four so3 molecules involved in the formation of the anhydride from the two pyrosulfonic acid precursors. As long as anhydride production is limited by short reaction . 8 times,. this error should be of limited significance. The pyrosulfonic acids, of course, are not expected to introduce any significant error into the analytical determinations since they are very active sulfonating agents themselves.36 The pyrosulfonic acids and the sulfonic acid anhydrides, insofar as they are sulfonating agents, may influence the overall rate of sulfonation~ This would affect the fit of the reaction to the selectivity 41 relationship. Were one able to isolate each of these secondary reactions and study them independently, it might

be expected that they would fit the lj,ne in Figure 6.

§ulfodehalogenation Sulfodehalogenation has been reported in the case of iodobenzene, based upon significant retention of radio-

activity when iodobenzene, sulfonated with 5350 3 was recrystallized with excess inactive benzenesulfonic acid. 72 That the separation of benzene- and substituted benzenesulfonic acids by recrystallization is not complete was ,. lateridemonstrated. 113 Titration of the products of· the sulfonation of chlorobenzene with Ag+ indicated insignificant dehalogenation.15 The possibility of debromination under the conditions of this investigation was s~udied. Bromobenzene was sulfonated by the usual method and the product converted to the sulfonyl chloride and separated by gas chromatography.lOS The chromatogram is reproduced in Figure 12~ The peaks, a, b, and d, obtained fr01n the separation were found to be b.romobenzene, benzenesulfonyl chloride and bromobenzenesulfonyl chloride, respectively. Since bromobenzene would not be carried along fran the' sulfonation step (being water

insoluble), it must be produced by desulfonation in the conversion to the sulfonyl chloride by the reaction with PCl5. The benzenesulfonyl chloride must result from the dehalogenation of bromobenzenesulfonic acid in either the 42

Fig. 12.--Chromatogram of the products.of the bromobenzenesulfonic acid conversion by phosphorus pentachloride. 43 sulfonation reaction or in th~ conversion to the sulfonyl chloride. Peak c was determined to be chlorobenzenesulfonyl chloride, apparently formed by the chlorodebromination of the bromobenzenesulfonic acid in the reaction with Pcl 5• Integration of these peaks gives the results shown in Table 12.

Table 12.--Products of the Conversion of Brornobenzene- sulfonic Acid to the Sulfonyl Chloride by Phosphorus Pentachloride - Peak Product Percent

a Bromobenzene 2.9 b Benzenesulfonyl Chloride o.s

C Chlorobenzenesulfonyl Chloride 0.4 d Bromobenzenesulfonyl Chloride 96.2

To check the previously reported deiodination, 72 iodobenzene was sulfonated by the same method and the sulfonyl chlorides produced and separated on the gas chromatograph. The results were almost identical, except for the retention time of iodobenzeneb"Ulfonyl chloride to those of the bromobenzene products, and are summarized in Table 13. The benzene solution of the iodobenzenesulfonyl chloride was a fairly intense red color, apparently fran the iodine produced but the chromatogram showed less than 1% benzenesulfonyl chloride. 44 Ta.ble 13.--Products of the Conversion of Iodobenzenesulfonic Acid to the Sulfonyl Chloride by Phosphorus Pentachloride

Peak Produe.t' Percent

a Iodobenzene 1.8 b Benzenesulfonyl. Chloride 0.7

C rCr1lorobenzenesulforiyl Chloride o.s· d Iodobenzenesulfonyl Chloride 96.6

Calculation of Relative Rates In order to calculate the partial rate factoz.:·o .for bromobenzene, it is necessary to know the relative rates for the reactions of benzene and·bromobenzene. This detennina- tion can be made most easily for fast reactions, su::h ·3.S sulfonations, as a competitive reaction and applying the Ingold equation 64 ,6S {37) to the resulting data.

log {¢Hi/¢Hf) (37) log (¢Bri/¢Brf)

where ¢H1 and ¢Br 1 are the initial concentrations of the aromatic compounds and ¢Hf and ¢Brf are the final concentrations. This equation is considered general for competitive reactions where the reaction is first order in 'both aromatics, although independent of t.~e order of the electrophile as long as the order is the same for both aromatic 45 compounds. The usual test for the Ingold equation is that k¢H/k¢Br is independent of the initial ratios of the substrate molecules. Several runs were ma.de .wherein mixtures of benzene and bromobenzene were sulfonated according to the procedure d~scribed in the experimental section. Several ratio$ of initial starting aromatics were studied, usually keeping the total moles constant. The experimental data for these runs are given in Table 14.

The a.mount of S03 could not be conveniently ~ontrolled so that various amounts appear in the different runs. The variation of the ratios of initial aromatic compounds covered a range of from 051 to 2.1. While there is considerable scatter in the data points, as noted in the benzene/chlorobenzene system,1 2 there is no significant correlation to be made between changes of relative rates and variations of starting ratios only, as noted in the other system, see Figure 13. Table 15 lists the relative rates in a descending order of the initial benzene/bromobenzene ratio. Table 14.--Experimental Data for the Relative Rate Determination of Benzene/Bromobenzene

Run Benzene Bromobenzene NaOH Sp .. Activity No. Grams Mmoles Grams Mrnoles Ratio Mls. Normality !'..moles cts./rng/min.

44 6.5487 83.84 13.4177 85.45 0.9812 · 15. 31 l.0274 15.73 235.7 45 6.5827 84.27 13.4450 85.63 0.9841 21.67 l.0274 22.64 248.2 46 6.5527 83.89 13.1697 83.87 1.0002 7.57 1.0274 7.77 230.9 47 6.6150 84.68 13.1203 83.56 1.0134 28.76 l.0274 29.55 264.5 48 6.6044 84.55 13.4106 85.41 0.9899 37.15 l.0274 38.17 234.3 49 s.so5a 112.74 9.0532 57.66 1.9553 35.00 1.0281 35.98 250.4 50 a.7657 112.22 9.0115 57.39 l.9554 28.82 l.0281 29.62 258.8 51 8.7307 111.77 8.6805 55.28 2.1888 16.76 1.0281 17.23 263.,8 52 4.4134 56.50 17.8417 113.63 0.4972 36.00 l.0281 37.01 208.,.l 53 4.4119 56.48 17.7414 112.99 0.4999 11.77 1.0281 · 12.10 251.9 54. 4.3825 56.ll 17.8393 113.61 0.4939 40.59 1.0303 41.82 213'.Jl: 55 4.4997 57.61 17 .,8344 ,113. 58 o.5012 2.19 1.0303 2.26 · 241.13 60 2.2478 28.78 35.6991 227.35 0.1266 27.33 1.0252 28.02 158~~ 61 2.2539 28.86 32.8843 209.43 0.1378 .14.64 1.0252 15.0l 179..:4 62 2.1990 28.16 37.3982 238.18 0.1182 38.22 1.0252 39.18 1331\:1 63 6.5504 83.86 13.3766 85.19 0.,9847 26.13 1.0265 26.83 230..:& 64 6.5480 83.83 13.5423 86.,25 o.9122 42.49 1.0265 43.61 243 ..•6 65 4.2777 54.76 17.7917 113.31 0.4837 34.37 l.0265 35.28 212.0 66 2.2137 28.34 35.5346 226.31 0.1252 5.89 1.0265 6.05 180.2 67 6.5104 83Q35 13.1873 83.97 o.9917 1.02;38 a.05 68 6.4634 82.75 13.3768 85.19 0.9707 8.91 1.0238 9.12 252.2 69 8.6945 111.31 8.7901 55.98 1.9875 6.89 1.0238 7.05 264.3 70 2.2082 28.27 37.1929 236.87 0.1193 61.86. 1.0238 63.33 98.8 71 6.7135 85.95 13.0935 83.39 1.0307 13.36 1.0238 13.68 .246. 2

~ 0\ 3.0 0 .&J '"IIS ~ c9 0 ·'-' 0 ~ 0 ,µ m C: 0 ~ 2.0 .µ i co as ~ ~ .µ 0 '"/tj 0 .... 0 Cl) ~ <9 0 0 0 1.0 I- Q)

0 ····.o.s ' 0 .s 1.0 1.5 2.0 Init:i.al Benzene/Bromobenzene Ratio

~ Fig. 13.--Apparent· relative rates as a function of initial benzene/bromobenzene ratios. ....; 48

Table 15.-~Relative Rates as Determined for Varying Initial Ratios of Benzene and Bromobenzene

Run Ratio . . _, k¢H/k¢Br No. ¢H/¢Br dl4 :--~--. G. c.

,51 2.1888 37.50 22.37 '69 l.9875 25.78 72.31 50 1.9554 , 21.35 17.09 49 1.9553 ll.47 11.90 :11. 1.0307 14.42 20.06 47 1.0134 56.75 1095.12 46 1.0002 .9.08 34.97 48 0.9899 12.34 17.12 63 0.9847 9.37 15.76 45 o.9841 19.80 39.01 44 0.9012 11.17 24.53 68 o.9101 19.72 43.73 55 o.so12 26.43 47,60 53 0.4999 46.87 36.54 52 0.4972 13.90 12.93 54 0.4939 10.,69 16.60 65 0.4837 14.76 18.07 61 0.1378 26.99 30~91 66 0.1252 6.62 57.90 62 0.1102 24.44 21.47

.. •· 49 Test for Thermodynamic vs. Kinetic Control While sulfonation reactions are COitll'llonlyreversible at relatively mild conditions, 89 it is not expected that this occurs under the conditions of this study. The· possibility of thermodynamic control rather than kinetic control for the final concentrations of the reactants and products was, however, considered. For a reversible reaction, /;

➔ + < (38)

the equilibrium constant would be obtained from

(39)

The data required for the calculation and the resulting equilibrium constants are given in Table 16. The data for both the c14 and gas chromatographic (G. c.) analyses are given. The wide range and lack of constancy indicate that the reaction is not thermodynamically controlled. Table 16.-Equilibrium Constant Calculation to Test Thermodynamic Control

Benzene Bromobenzene Hydrocarbon Sulfonic Acid Hydrocarbon Sulfonic Acid Run Keq No; c14 G. C. cl4 G. C. cl4 G. C. cl4 G. C. cI4 G. C.

44 69.53 68.80 14.31 15.04 84.03 84.76 1.42 0.69 0.602 0.037 45 62.89 62.69 21.38 21.98 84.37 84.97 1.26 0.65 0.044 0.022 46 76.92 76.35 6.97 7.54 83.07 83.64 o.ao 0.23 0.103 0.027 47 55.50 55.16 29.18 29.52 83.19 83.53 0.37 0.03 o.ooa 0.0004 48 49.94 49.06 34.61 35.49 81.85 82.73 3.56 2.68 0.063 0.045 49 78.55 78.49 34.19 34.25 55.87 55.93 1.79 1.73 0.074 0.071 50 83.39 83.59 28.83 28.64 56.60 56.41 o.79 0.913 Oec04_0 0.051 51 94.78 94.90 16.99 16.87 55.04 56.92 0.24 0.36 o.02~ 0.036 52 25.74 26.08 30.76 30.42 107.38 l07a04 6.25 6.59 0.0~9 0.053 53 44.93 45.08 11.55 11.40 112.44 112.29 0.55 0.70 O.OL9 0.025 54 20.75 20.86 35 •.36 35.25 107.15 107.09 6.46 6.57 0.(?35 0.036 55 55.51 54.44 2.10 2.17 1-13.42 113.49 '0.16 0.09 O.Olfi 0.002 60 9.76 9.33 19.02 19.45 218.35 218.78 9.00 8.57 o.on 0.019 61 17.64 17.30 11.22 11.56 205.06 205.98 3.79 3.45 0.029 0.025 62 s.10 5.83 23.06 22.33 222.06 221.33 16.12 16.87 0.01'6 0.020

UI 0 51 Analysis of the Data for Secondary ReactionJ!_

It is quite obv.i~t .,f:tdfi t.hl data presented in Tables 17 and 18 that other factors besides the starting ratios of benzene and bromoben~ene are very significant in controlling the relative rates of these sulfonation reactions. This same observation was made in the benzene/ chlorobenzene case. The ratio of the products, benzenesulfonic/chlorobenzenesulfonic acids was found to be strongly dependent upon the so3 concentration--the ratio decreased with increasing so3 .concentration. Plots of the dat~ in Tables 17 and l87as shown in Figures 14 and 15, show '· the same general behaviour as the corresponding plots of the benzene/chlorobenzene system. The results of four benzene/ chlorobenzene sulfonations in a ratio of 0.129 gave an approximately horizontal and straight line when plotted against the S03 concentration'. The benzene/bromobenzene ratio of 0.127 gives a fairly straight line with _a slight slope. The curvature of each of the other lines appear to approach a horizontal region in the limit which represents a concentration of s03 above which the ratio of the sulfonic acid products is unaffected by further excesses of S03. This area, then, represents a region where the secondary reactions, probably the sulfonation by the pyrosulfonic acids and the sulfonic acid anhydrides, are sufficiently ~ore-important than the primary reaction that the product distribution of the primary reaction is completely Table 17.--Relative Rate Determination Utilizing Gas Chromatographic Separation of the Sulfonyl Chlorides

Benzene Bromobenzene Moles1 Mi Moles Run Mmoles sulfonate :-1moles Log- .Mmoles sulfonate Mrnoles 1 Rel.Rate No. Init. Final Molesf Init • Final Log!:!!. M%1Mrnoles Mf M%JMrnoles Molesf Mf k¢f'k¢Br

44 83.83 95.6 15.04 68.80 1.2186 0.08586 85.45 4.4 0.69 84.76 1.0081 0.00350 25.53 45 84.27 97.1 22.01 62.26 1.3535 .13146 85.63 2.9 0.66 84.97 1.0078 .00337 39.0l 46 83.89 97.l 7.54 76.35 l.0988 .04092 83.87 2.9 0.23 83.64 1.0021 .00117 34.97 47 84.68 99.9 29,52 55.16 1.5352 .18617 83.56 0.1 0,03 83.53 1.0004 .00017 1095.12 48 84.55 93.0 35.50 49.05 1.7238 .23649 85.41 7.0 2.67 82.74 1.0323 .01381, 17el2 49 112.74 95.2 34.25 78.49 1.4364 .15727 57.66 4.8 1.73 -55.93 1.0309 .Olm' 11.90 so 112.22 96.7 28.64 . 83. 58 1.3427 .12748 57.39 3.3 0.98 56.,41 1.0174 .007llJ~. 17.09 51 111.77 97.9 16.87 94 ..90 1.1778 .06108 57 .. 28 2a 0.36 56.92 l.0063 .0027~· · 22.37 52 56,..S0 82.2 30 ..42 26.08 2.1664 .33577 113.63 17.8 6.59 107.04 1.0616 .• 02$96 12. 93 . 53 5.6.48 94.2 11.40 45.08 1.2529 .09792 112.99 5.8 0.10. J..12.29 l.0062 .002:tiik 36.54 54 56.ll 84.3 35.25 20.86 2.6898 .4297.2 113.61 15.7 6.57 107.04 1.0614 .02~e 16.60 55 57.61 96.l 2.11 55.44 1.0391 .01666 113.58 3.9 0.09 113.49 1.0008 .0083'$ 47.60 60 28.78 69.4 19.45 9.33 3.0847 .48922 227.35 30.6 8.57 218.78 1.0392 .Olb-lt:i 29.29 61 28.86 77.0 11.56 17.30 1.6682 .22225 209.43 23.0 3.45 205.98 l.0167 .00719 30.91 62 28.16 57.0 22.33 5.83 4.8302 .68399 238.18 42.0 16.85 221.33 1.0761 • 03185 21.47 . 63 83.86 92.9 26.78 57.08 1.4692 .16708 85.19 7.1 2.05 83.14 1.0247 .01060 15.76 64 83.83 93.0 40.57 43.26 1.9378 .28731 86.25 7.0 3.05 83.20 1.0367 .01565 18.36 65 54.76 85.4 30.13 24.63 2.2233 .34700 113.31 14.6 5.15 108.16 1.0476 .01920 18.07 66 28.34 83.9 5.08 23.26 1.2814 .10769 226.31 16.l 0.97 225.34 1.0043 .00186 57.90 67 · 83.35 97.6 7.86 75.49 1.1041 .04301 83.97 2.4 0.19 83.78 1.0023 .00100 '43.01 68 82.75 97.6 8,90 73.85 1.1205 .04942 85.19 2~4 0.22 84.97 1.0026 .00113 43.73 69 111.31 99.3 7.00 104.31 1.0671 .02820 55 .. 98 0.7 o.os 55.93 1.000~ .00039 72.31 70· 28 .. 27 45.7 28.94 236.87 54.3 34.39 202.48 1.1698 .06811 71 85.95 95.0 13.,00 72.95 1.1782 .071-22 83.59 5.o· 0.68 82.91 1.0082 .00355 20.06

U'I ti,) Table 18.--Relative Rate Determination Utilizing Labeled Benzene

Benzene Bromobenze.ne ::r.: tr! rt) rt) ::r.: 0 (I) 0 (Y) Molesi Cl) Q) CJ) Moles Rel.Rate Run Moles Moles Moles Lo ~ Moles r-1 'Q Moles 1 Log~ 0 1-t No. Init. ~fJt,'\"Q ,¢S03H Final Molesf g Mf Init. Final MoleSf Mf k,Vk¢Br ~! ~r:Q 44 83.84 87.5 14.23 69.51 1.2062 0.08142 85.45 12.s 1.42 84.03 l.0169 0.00729 11.17 45 84.27 92.2 21.38 62.89 1.3400 .12710 85.63 7.8 1.26 84.37 1.0149 .00642 19.,80 46 83.89 as.a 6.97 76.92 1.0906 .03767 83.87 14.2 o.ao 83.07 1.0096 .00415 9.08 47 84.68 98.2 28.94 55.74 1.5192 .18160 83.56 1.8 0.61 82.95 1.0074 .00320 ~ 56. 75 48 84.55 87.1 34 .. 60 49.95 1.6927 .22858 85.41 12.9 3.57 81.84 1.0436 • 01853-, Ii' 12.34 · 49 112.74 93.0 34.19 78.55 1.4353 .15694 57.66 7.9 l.79 55.87 1.0320 .01368,:,. 11.47 50 112.22 96.2 28.83 83.39 1.3457 .12895 57.39' 3.8 0.79 56.60 1.0140 .00604·~. 21.35 51 111.77 98.0 16e99 94.78 1.1793 .07162 55.28 2.0 ~o.24 55.04 1.0044 .00191" 37.50 52 56.50 77.4 30.76 25.,74 2.1950 .34143 113.63 22.6 6.25 107.38 .1.0582 ~0245~ '= 13. 90 53 56.48 93.6 11.55 44.93 1.2571 .09937 112.99 6.4 0.55 112.44 1.0049 .00212 46.87 ...r ' 54 56.ll 78.2 35.29 20.82 2.6950 ..43056 113.61 20.0 10.06 103.55 1~0972 • 04029-1 ;- 10. 69 55 57.61 89.9 2.10 55.51 1 ..0378 .01612 113.58 10.1 0.16 113.42 1.0014 .0006};;;!,,\ 26.43 60 28.78 59.5 19.02 9.76 2.9488 .01612 227.35 40.5 9.00 218.35 1.0412 .01753, 26.79 61 28.86 67.3 11.22 17.64 1.6360 .21378 209.43 32.7 3.79 205.64 1.0184 .00792 26.99 62 28.16 49.9 23.07 5.09 s. 5,324 .74291 238.18 50.l 16.11 222.07 le0725 .03040 24.44 63 83.86 84.9 23.87 59.99 1.3929 .14392 85.19 15.1 2.96 82.23 1.0360 .01536 9.37 64 83.83 86.3 39.28 44.55 1.8817 .27455 86.25 13.7 4.34 81.91 1,.0530 .02243 12.24 65 54.76 78.0 29.49 25.27 2.1670 .33586 113.31 22.0 5.79 107.52 1.0538 .02276 14.76 66 28.34 66.3 4.47 23.87 1.1873 .07456 226.31 33.7 o.58 22.05 1.0263 .01127 6.62 67 9'3. 35 -~- 83.97 ' 68 82.75 92.8 8.65 74.iO 1.1167 .04793 85.19 7.2 0.47 84.72 1.0056 .00243 19.72 69 111.31 97.2 6.91 104.40 1.0662 .02784 55.98 2.8 0.14 55.84 1.0025 .00108 25.78 70 28. 2.7 36.3 28 •.52 236.87 63.7 34.81 202.06 l.J,723 .06904 71 ·a5.95 90.6 12.76 73.19 1.1743 .06978 83.39 9.4 0.92 82.47 1.0112 .00484 14.42 ' - UI w 54 50 - \ Initial Ratios -:.; ¢H/¢Br = 2.0 Q ¢H/¢Br = 1.0 +\6. ¢H/¢Br = o.s D ¢H/¢Br = 0.1 40 0

30 -,-I

tO -~ 'O \ a u•

(.?• 0 20 · -.-I . -tr: P') 0 0 \

i1-t 6

sP') 0 0 ~ 10

'□ 'I:!,.~ ~ . .b. □------.. □---' □- 0 10 20 30 . 40 S03 mmoles

Fig. ).4.--variations in• the ratio c,f the product sulfonic acids with sulfur trioxide concentrations as dete~ined by gas chromatography. 55 -,-' Initial Ratios -,-I ¢H/¢Br = 2.0 ;: 60 Q ¢H/¢Br = 1.0 ¢H/¢Br \'"f, D,, = ·o.s 0 ¢H/¢Br = 0.1

50 -,-I

RS -+' «s 't1 ~ ....u 30

-~ P') 0 tVl A 20 I sP') 0 -,- {I) 0 \ "Q

10 ' 0 ~ 0 -0 A~ ------□ . ~------□----- □-

0 10 20 30 40 S03 mmoles

Fig. 15.--variations in the ratio of the product sulfonic acids with sulfur trioxide concentrations as determined by c14 analysis. 56 overshadowed. The appearance of the plots suggests the possibility that all curves "'will eventually level off at about the same benzenesulfonic aaidfbromobenzenesulfonic acid ratio for all initial ben2ene/bromobenzene ratios. This horizontal portion of the line would indicate that the ratio of the products from the primary reaction is the same as the ratio of the products from the secondary reaction. The partial rate factors for the secondary reaction might possibly be determined if the isomer distribution of the substituted benzene were determined at a concentration of S03 high enough to correspond to the same region. , Brown extrapolated the products of the sulfonation of benzene/chlorobenzene mixtures to a zero sulfur trioxide concentration suggesting that this would give the relative rates of the primary reaction under conditions free of secondary effects. This is n?t necessarily the case. If secondary reactions are actually involved, as opposed to side reactions, then, according to the system described on page 28, ~~•,the early, linear, portion of the plot in Figures 16 and 17 merely represents an area where changes in the relative concentrations of the two aromatic compounds are not sufficient to give a significan~ deviation. ,only if \ the reactions involved in the sulfonation were comp,titive, not secondary, would j_t be possible to separate the two kinds of reaction. It is readily seen from.Figures 16 and 17 that for each initial ratio of starting aromatic compounds, there Table 19.-Moles of the Final Products in the Competitive Sulfonation of Benzene and Bromobenzene

Labeled Benzene G. c. Separation Run Initial ¢s03H .· Br¢S03H ¢H03H/BrS03H ¢so3H Br{6S03H ¢s03H/Br¢s03 S03 No •. Ratio Mm.oles Mmoles Ratio Mmoles Mmoles Ratio Mmoles 44 0.9812 14.23 1.42 10.02 15~05 0.10 21.50 15.75 45 0.9841 21.38 1.26 16.97 22.01 0.66 33.30 22.64 46 1.0002 6.97 o.ao 8.71 -7 .. 54 0.23 32.78 7.77 47 1.0134 28.94 0.61 47.44 29#52 0.03 984.00 29.55 48 0.9899 34.60 3.57 9.69 35.50 2.67 13.24 38.17 68 0.9707 8.65 0.47 18.40 8.90 0.22 40.45 9.12 71 l.0307 12.76 0.92 13.87 13.00 0.68 19.12 13.68 Avga0.9957 49, 1.9553 34.19 -1. 79 i9.l0 34.25 1.73 i9.80 -35_ 98 50 1.9554 28.83 o.79 36.49 28.64 0.98 29.22 .29.26 51 .,., 2.1888 16.99 0.24 70.79 16~·87 · 0.36 46.86 . -17. 23 69 l.9875 6.91 0.14 49.36 7.00 o.os 140.00 .,:·7.05 Avgs 2.0218 52 0.4972 30.76 6.25 4.92 30.42 6.59 4.62 37.01 53 0,.4999 11.55 o.ss 21.04 11.40 0.10 16.29 12.10 54 0.4939 35.29 10.06 3.51 35.25 6.57 5.37 41.82 55 0.50.72 .2,10 0.16 13 .. 12 ... 2.17 0.09 24.11 2.26 65 0.4837 29.49 5.79 5.09 30.13 5.15 5.85 35.28 Avg1 0.4964 ' 60 0.1266 19.02 9.00 2.11 19.45 .,:' 8.57 2.27 28.02 61 0.1378 11.22 . 3. 79 2.96 11.56 3.45 3.35 15.01 62 0.1182 · 23.07 16 ..ll · 1.43 22.33 ,,. -116.85 '·l.33 39.18 66. 0.1252 4.47 o.sa . 7.07 5.08 0.97 s •.24 6.05 Avg1 0.1270 i UI ...J Initial Benzene/Bromobenzene Ratios 30 6.--6-- -1-¢H/¢Br = 2.0 ¢H/¢Br 1.0 0 = ,,-;

6.., ¢H/¢Br = ~:•.o.s 25 D ¢H/~Br = 0.1 ~□- 20 !ll □ II) /0 / r-1 /-.-I D 15 ~ ::c:i /0 •·.. M gl /0 ~ 10 /'"' .,,o 0 . o' D A-~- ~~A .. 5 o' / . D I 0 A~ ~ I ------•,• --- .o ;.. AO - o··- • - -!,r."\-l J-:----1,j- . -I--=..- 0 5 10 15 20 25 30 35 40 45 50 Mmoles so3 U'I Fig. 16.--Moles of sulfonic acid products/sulfur trioxide (G. c. data). co Initial Benzene/Bromobenzene Ratios 30 I- / ~ Ll. -- -:- ¢H/¢Br = 2.0 O ¢a/¢Br = 1.0 6. f6H/(6Br o.s ..25 I- = D ¢a/¢Br = 0.1 ,,,, , □- ~ : 20~ 0 r-4 /° 1/ g -.- / I / 0 : 15 / ('I) 0 (/) ~ L.l w 10 i / /. o/ ~

- - - 6/ 5 ---- -:--- 0 15 20 25 30 35 40 45 50 S03 mmoles

14 U1 Fig. 17.--Moles of .product sulfonic acid ·vs. sulfur trioxide (c data). t..O 60

is an S03 concenti-:ation bel:ow which the ratio of the products is linear. While. as indicated above, this does not necessarily mean that the prima~y reaction is dominant. this is a region where the products are fairly "standard." For the reactions with initial benzene/bromobenzene ratios of l.O or greater, nearly straight lines were obtained

throughout the range of so3 concentrations used. Brown never plotted any data in this manner for 'Which the initial ratios of benzene/chlorobenzene were greater than 0.708. It is obvious from his data. however, that at the higher initial ratio. the plots of the product sulfonic acids became more linear.

From straight lines drawn from the data point~ of the 2.0 ratio. a relative rate constant ratio of 28.6 is calculated for the comoetitive reaction of ·benzene and - "~. bromobenzene in reaction with sulfur trioxide in. liquid sulfur dioxide at -12 to -13°c.

Isomer Distribution for the Sulfonation of Bromobenzene Reactions intended for detennining the isomer distribution of the products of the sulfonation of bromobenzene were carried out in the same manner as the competitive sulfonation of benzene/bromobenzene mixtures 35 except that s labeled so3 was used and the bromobenzene was not mixed with benzene. The work-up of the r-eaction products was the same to the point of titration with standardized NaOH. After this step, aliquots of this 61 titrated solution of the product sulfonic acids were added to l.arge quantities (about 40 g.) of the three previously prepared isomerically pure inactive bromobenzenesulfonic . -·~ ,,.-; ·" , .. acid salts. To these were added a calculated 5% excess of HCl and p-toluidine and each isomer was repeatedly recrystallized as the p-toluidine salt with 400-500 mg. samples of each recrystallization being reserved for future counting purposes. Each isomer was recrystallized as long as sufficient quantities of the salt remained. Usually thirteen to twenty-one recrystallizations were obtained. An initial group of five runs gave consistent results for the ortho- and para-isomers, but the activity of the meta isomer was found to fall to a minimum after about five recrystallizations; however, instead ~f leveling off as expected, the activity was found to increase often to twice or three times the initial activity. This was probably due to the purification of the para isomer instead of the meta in spite of the large excess of the meta isomer. Seven new runs were made later using the same technique except that the salts were allowed to recrystallize more slowly and no effort was ma.de to obtain maximum crystallization (no ice bath was utilized). These runs gave data for the meta isomer that were consistent with the ortho and para isomers. The first of these seven gave results a little off from the rest, which wer:e counted in a freshly prepared Kinard solution. The counting data are given in Table 20. 62

Table 20.--Counting Data for Isomer Distribution Determination

Recryst. No. Para Ortho Meta

34-s 1 177.2 298.6 138.7 3 174.8 16.7 30.7 5 174.9 9.0 24 •.3 7 173.9 8.6 24.3 9 171.S a.9 10.3 11 173.4 8.7 10.s 13 175.3 8.6 a.s 15 172.2 8.3 11.a 17 175.0 9.3 a.o 19 168.4 21 170.6 Averages 173.0 - a.a 9.9

35-s 1 120.2 168.9 230.9 3 118.7 17.3 9.9 5 119.2 13.2 3.5 7 119.6 12.4 3.2 9 119.2 12.a 3.1 11 118.5 12. 7 3,3 13 118.2 12.9 3.1 15 119.0_ 11.5 3.2 17 121.a 11.5 3.2 19 3.0 21 3.1 23 3.2 Average: 119.0 12.5 3.2

36-s 1 131.2 · 194.3 178.0 3 126.6 15.2 s.s 5. 130.3 12.6 2.4 1·· 127.0 12.4 2.2 9 126.8 12.s 2.3 11 125.1 12.6 2.2 13 122.9 12.7 2.2 15 123.7 2.1 17 121.1 2.1 Averages 124.4 12.6 2.2

! 63 Table 20 (continued)

Recryst. No. Pltj:,a ortho Meta

37-s l 134.3 l.70.4 104.5 3· 132.9 13.8 4.1 5 132.6 12.7 2.9 7 129.9 11.9 .2.4 9 132.0 11.8 2.4 11 133.6 12.0 2.4 13 131.7 11.8 2.s 15 130.8 11.9 2 .•4 Average: 132.0 11.9 2.4 38-S l 192.0 521 .. 6 505.4 3 187.1 26.4 20.2 5 188.5 15.7 3.7 7 187.9 15.7 3.3 9 189.6 15.7 3.i 11 188 .• 7 15.6 3.1 13 187.2 15.8 3.1 15 189.7 15.5 3.1 Averagea 188.4 15.7 3.2

39-S l 275.3 1052.9 528.8 3 267.4 27.8 18.3 5 266.8 27.7 5.3 7 269.9 27.1 5.0 9 266.l 27.7 4.9 11 268.2 27.3 4.9 13 270.5 27.3 4.8 15 268.6 27.1 s.2 Avera.ge: 268.2 27.4 5.()

40~s 1 58.1 64.S ~ 300.9 3 56.6 5.3 20.2 5 56.l 4.4 1.3 7 56.l 4.5 0.1 9 55.7 4.3 o.a 11 56.6 4.4 O·.9' 13 55.7 4.4 o.9 15 56.4 4.3 o.s Average: 56.2 4.4 o·.0 64 T"ne percent of ~ach isomer in each run was calculated from the ratios of the aliquot volume added to each inactive salt, the ratio of.the moles of inactive salt used and the specific activity determined in the· counting data above. The isomer distributions thus calculated are given in Table 21.

'!'able 21.--Isomer Distribution of Bromobenzenesulfonic Acid

Run % Para % Ortho % Meta ml. ¢Br g. S03 used

34-S 97.98 1.01 1.01 10.0 1.554 35-S 98.72 1.04 0.24 10.0 2.040 36-S 98.87 0.97 0.16 10.0 1.321 37-S -98.94 0.90 0.16 10.0 1.338 38-S 99.04 0.82 0.1.s 10.0 1.825 39-s 98.88 o.96 0.16 s.o 5.763 40-S 99.11 o.76 0.13 20.0 0.683

These results are more consistent with the isomer distributions previously determined for chlorobenzene and iodobenzene while being significantly different than previously reported for bromobenzene. The earlier work on bromobenzene involved counting the radioactive salt on planchets in a Geiger counter, which is less accurate than the scintillation counting method used in this work. These results are summarized in Table 22. 65 Table 22.--The Isomer Distributions in the Sulfonation of the Halobenzenes

" Halobenzene % ortho ; "M!ti. % Para Reference Chlorobenzene 0.95 ± 0.03 Q.09 ± 0.02 98.96 ± 0.12 13 Bromobenzene 0.6 :I: 0.2 '3.6 ± 0.0 95.9 ± 0.6 107 Bromobenzene 0.91 ±. 0.04 0.1s :t 0.02 98.93- ± o.14 This work Iodobenzene 1.40 ± 0.11 0.34 ± 0.11 98.26,± 0.11 112

While this study lacks sufficient data to show conclusively the differences in the isomer distribution with concentrations of sulfur trioxide~ the effects of secondai·y reactions are not obvious •. The ortho/para ratios for runs 34 through 40 are shown in Table 23.

Table 23.--ortho/Para Ratios for the Sulfonation of Bromobenzene

Percentages Run S03 Bromobenzene Ortho/Para No. mmoles ml. ortho Para Ratio

34 19.41 10 1.01 97.98 0.0103 35 25.48 10 1.04 98.72 o.01os- 36 16.50 10 0.97 98.87 o.009a 37 16.71 10 o.9o 98.94 0.0091 38 22.79 10 0.82 99.04 0.0083

39 71.97 5 0.96 98"88 0.0097 40 8.53 20 0.76 99.91 0.0011 66 The five isomer distribution runs which gave anomalous results for the meta isomer, runs 20-24, did give consistent results for t:hl ~Qr,thci,•nt para isomers. The ' ortho/para r~tios for these runs are calculated and shown in Table 24 and do not show the same behaviour as runs 34-40. The earlier runs were made in 765 ml. total volumes in liquid sulfur dioxide, while runs 34-40 were made in total ,_. volumes of 635 ml. The differences arising from the difference in concentration are in the direction that would !i be predicted on the basis of secondary reactions.

Table 24.--ortho/Para Ratios for Runs 20 through 24

Specific Activity Run S03 Bromobenzene Ortho/Para No. mmoles ml. Ortho Para Ratio 20 23.26" 10 11.09 862.l 0.0138 21 23.84 10 13.24 954.3 0.0139 22 22.32 10 11.23 810.0 0.0139

23 8.16 20 5.27 363.2 0.0145

24 34.22 5 26.85 2179.4 0.0125

Calculation of Partial Rate Factors From the data obtained in this study, it is now possible to calculate the partial rate factors f9.r the

sulfonation of bromobenzene with so 3 in liquid s02 at -12°c. The relevant data are given in Table 25, along with the values necessary for the reactivity-selectivity relationship. These are plotted in Figure 6. 67 Table 25.--PartialRate Factors for the Sulfonatiori of Bromobenzene with S03 in s02

Term ·Value

% ,2- 9Ef~g3 ! .14 % £- 0.91 * .04 % m- 0.17 z .02 Pf 0.208 Of 0.0096

mf 0.00018 pf/~ 1163.9 log pf/mf 3.065 log Pf -0.682

It was suggested by Brown that, at low concenti:·ations

of S03, the primary reactions predominate over the secondary reactions. It would follow, then, that the isomer distribution obtained at the lowest S03 concentration should give a better fit to the reactivity-selectivity line predicted from op+ and tTm+ values. Utilizing the isomer distribution data for run 40-S for the calculation of

partial rate factors, gives the values in Table 26 and,,_ point 30 in Figure 6. It is seen that this point is further removed from the predicted line than when the average isomer distribution of the higher S03 concentrations was used. 68 Table 26.--Isomer Distribution and Partial Rate Factors for Low S03 concentration

Term Value

% p- ,gb.11

"2- 0.76 .. %~ 0.13 Pf 0.204 of 0.0078 ffit 0.00013 Pflmf 1522.4 log Pf/Inf 3.182 log Pf ·o.6BO

Error Analysis Since each determination of the relative rate was, in fact, an individual measurement, these data could not be analyzed by the standard methods for accuracy and precision. The precision of the measurements and determinations required for the calculation of the relative rates is indicated in Table 27. The initial weight of each hydrocarbon was not precisely determinable since the method of addition to the reaction mixture involved decanting without consideration of the small amounts retained in the weighing flask. By the difference in the·weight of the flask before weighing out the hydrocarbons and after decantation, :I.t was determined that the decantation was 99.97 to 99.98% complete. 69 Table 27.--Precision of the Measurements Used in the Calculation of Relative Rates == Determinations Error Limits

Weight of Initial Beitz~- :' ~ , ·.• • 0 • 0 2 g •

Weight of Initial Bromobenzene • • • 0.02 g. Mole Ratio of Benzene/Bromobenzene ~ 0.0002

Mmoles of sulfur Trioxide • • • • • 0.001 mmoles Normality of Base • • • • • • • • • 0.0005 mrnoles/ml. Volume of Base • • • • • • • • • • • 0.02 ml. Weight of Counting Samples • • • • • 0.0002 g • Background Count Rate ••••••• 5 cts./min. Standard Count Rate • •' • • • • • • 1.2 cts./min./mg ...

Integration of Chromatogram • • • • 2% - 5%

The ratios of the aromatic compounds in the initial mixtures were known with reasonable accuracy, however. The error limits of standard measurements ·indicated in Table 27 are generally greater than is normally considered probable and when taken additively would account for deviations of about± 0.40 units of kca/kiBr• While the accuracy of the G. c. analysis of the ., product sulfonyl chlorides inherently involves variations in .. the sensitivity of the detector towards the various products, the limiting factor in this analysis was undoubtedly in the integration. cutting out the peaks and weighing on an analytical balance is an accepted method of integration; 70 however, especially with small peaks, the error in cutting out may approach several percent, on larger·peaks, about 2% seems·reasonable.

,. Wher~ multiple determinations were made, as with the isomer distribution determination, the standard deviation was determined and is reported in a-units.

(J· = (40) where (a-x) is the deviation from the average of n determinations. CHAPTERIV

EXPERIMENTALSECTION

Preparation of the Isomers of Bromobenzenesulfonic Acid The isotope dilution' technique was used to determine the isomer distribution of the bromobenzenesulfonic·'acids produced in the sulfonation of bromobenzene. · Quantities of each isomer were prepared by a method which precluded the formation of other isomers according to Meerwein, ~ al. 78 The steps involved in the syntheses were the diazotization of the respective bromoanilines, decomposition of the diazonium chloride with so2 in acetic acid, and hydrolysis of the product bromobenzenesulfonyl chloride. Potassium

NH2 N2Cl S02Cl

NaN02 S02 ·-:acl ► HOAc Cu2Cl2 r so3 - (41)

KOH ) J<"f" H20

hydroxide was used in the hydrolysis of the sulfonyl chlorides •. The potassium salt of the meta isomer was not

71 72 conveniently recrystallizable, so the potassium salt was converted to the bariu.~ salt of m-bromobenzenesulfonic acid by the addition of barium chloride to the solution of potassium m-bromobenzenesulfonate.

A typical pr.eparative·run involved dissolving 200 g • .. (1.16 moles) of p-bromoaniline C~Udrich Chemical Co.) in 375 ml. of ethyl ether and, while maintaining the temperature at less than 10°c, adding 410 ml. of concentrated hydrochloric acid. To the p-bromoaniline hydrochloride was then added with cooling and vigorous stirring 85 g. {1.3 moles) NaNo2 in 175 ml. of water. During thi~ addition, 900 ml._of glacial acetic acid, containing 37 g. of CU2Cl2•2H20 (Baker's c. P. grade) dissolved in a minimum amount of water, was saturated with sulfur dioxide (war surplus, grade unknown) in an 8-1.jar. The diazoniu.~ chloride was poured into the chilled acetic acid-sulfur dioxide solution as rapidly as possible with vigorous stirring and/or use of a compressed air stream. This was necessary to control the profuse foaming and prevent overflowing of the container as the nitrogen gas was released in the displacement of the diazo group. Bubbling of s02 through the reaction mixture was continued for an hour. The g_- and _e-·bromobenzenesulfonyl chlorides solidified upon addition of 1500 ml. of crushed ice to the sulfonation products. 'l'hes.e were collected by filtration 73 and washed with cold water and hydrolyzed with potassium hydroxide (Allied Chemical, reagent). The m-bromobenzebertulfotvt dhloride was a liquid and - . - :. 1-' < was separated after the addition of 1500 ml. of "crushed ice from the reaction mixture by extraction with five-300 ml. portions of ethyl ether. The ether was boiled off and the sulfonyl chloride was hydrolyzed.with potassium hydroxide.

The potassium salt was converted to the barium,salt by the addition of barium chloride to the solution of the potassium sa.lt. The solutions of each of the bromobenzenesulfonic acid salts were treated with Norit and filtered and crystallized readily when coolede The salts were collected by vacuum filtration and recrystallized from a 75/25 water- ethanol solvent system until pure. The overall.yields for the three isomers were 49% of theoretical for the meta isomer, 59% for the para and 51% for the ortho isomers. The s-benzylisothiouronium derivatives32 of the isomeric bromobenzenesulfonic acid salts were prepared by dissolving 0.25 g. of the salt in water and adding l ml. of an aqueous 10% s-benzylisothiouronium chloride solution. The product was recrystallized fr;_pmwater three or four times. The melting points of the s-benzylisothio,uronium bromobenzenesulfonates are given in Table 28. 74 Table 28.--Melting Points of the s-Benzylisothiouronium Salts of the Bromobenzenesulfoilic Acids

Isomer ~it,coc, Reference ortho 147-148 148.5-151 108

Meta 130. 5-131. 5 . 132-133 108

Para 175.5-177 170 33

177-178 108

Preparation of Radioactive sulfur Trioxide 35 The s35 was received as a2s o4 from New England Nuclear Corporation. This was washed into a beaker containing 2.0 g. of BaS04 (Baker and Adamson, reagent) and 0.25 g. of BaCl2 (Brothers Chemical co., reagent). This mixture was then digested on a steain plate for several days. The resulting Bas35o 4 was washed thoroughly and the water decanted three times and dried overnight under vacuum at 11s0 c. 35 ·.The Bas o4 was then divided up between four exchange reactors, Figure 18, which were then loaded with 4 to 8 ml. of S03 (Sulfan B, Baker and Adamson, stabilized). The reactors were immersed in an.ice water bath to freeze the so3 and then evacuated to about l torr and sealed off at the constriction, B. Each exchange reactor was kept on a steam plate for several days with occasional shaking for the exchange to 75 equilibrate. The S03 was repeatedly distilled from the exchange chamber, A, into section C which was thoroughly rinsed along with the thin-walled capsules, D, to remove any 0 traces of water and sulfuric, a~ id. The capsules were individually filled with S03, frozen and sealed off for use.

A Fig. 18.-Isotope exchange react.or.

Capsules containing non-radioactive so3 were prepared in a similar manner exclusive of the barium sulfate and exchange process for use in the competitive sulfonation of benzene-bromobenzene mixtures.

The amount of so3 varied between capsules, but the amount contained in any given capsule was determined by the titration of the sulfonated products with standardized

sodium hydroxide solution. A check of the amount of so3 in each capsule was made by weighing the filled capsule before use and then collecting the:fr~gments after the sulfonation and subtracting their weight fran that of the filled capsule. 76 Preparation of Radioactive Benzene ~ Radioactive benzene, labeled with cl4 was obtained from New England Nuclear Corporation~· It was received with an activity of 0.10 me. in a meak-seal vial. This was diluted to 300 ml. after removal from the break-seal vial. This vial was sealed to a vacuum line, Figure 19, connecting it with a Erlenmeyer flask containing 25 ml. of pure ~..nzene. With the benzene frozen, by immersing the flask in liquid nitrogen, the system wa.s evacuated and the·sealed vial opened by shaking the piece of glass rod against the break seal. The yial and the Erlenmeyer were alternately cooled in liquid nitrogen and warmed with a heat gun until thorough mixing of the benzene occurred and, finally, complete transfer of the benzene from the vial effected.

Fig. 19.--vacuum line for c14 benzene preparation. 77 Purification of Benzene and Bromobenze..!!!, The benzene (reagen~ grade) was predried over sodium metal and then distilled through a 30-plate (4-ft. glass helices-packed) vacuum jacketed co,lumn fitted with a total- condensing, partial-take-off head. A reflux ratio of 1011 to 2011 was used. A center cut of about one-half the total was collected and stored in a tightly stoppered bottle in the refrigerator. The bromobenzene (reagent grade) was distilled,

collected and stored in the same manner. Samples of the benzene·. and branobenzene were

injected into a Varian model 200 gas chromatograph. No

.contamination was apparent at the maximum sensitivity of the instrument.

Sulfonation Procedure

The sulfonation,., reactions,r were conducted in the same apparatus for both the isomer distribution experiments and the competitive rate experiments. In the former 35 case s o3 and non-radioactive bromobenzene was used·while,

in the latte~ case, a mixture of non-radioactive so 3 and c14- l~beled benzene and bromobenzene were used. The apparatus, Figure 20, consisted essentially of a one-liter,

three-necked main reaction vessel (A) fitted with a three

junction copper-constantan thermocouple (B) and a siphon tube connected to a smaller condensing chamber (C) and cold-

finger assembly (D). The syst~ was vented t.hrough a. mercury bubbler and a drying column to protect the system 78

Gases in --➔ /}_ ___

D I

Gases out u 1 E d

Fig. 20.--The sulfonation apparatus. 79 against moist air which might occasionally-have been sucked back. .. All of the gases, so2 (Matheson, anhydrous grade) and ,} ' N2 (99%, anhydrous) were passed through a drying tube (P20s supported on glass beads) prior to introduction into the system .. The apparatus was carefully dried in a drying oven at 110°c for at least 8 hours between runs. It was then assembled while warm and dry nitrogen was flushed through the system during and for 15 minutes after ·assembly, prior to the initiation of a run. - After assembly and flushing with the dry nitrogen, the main reaction ch:1mber was surrounded with a lal mixture of chloroform and carbon tetrachloride and cooled to -20° to -40°c with dry ice. In a similar way the condensing chamber iras cooled and the cold-finger filled with methanol and dry ice., sulfur dioxide was introduced a11d condonsed on the cold finger and collected in the condensing chamber. About 100 ml. was collected and passed over into the main reaction chamber, via the siphon tube by closing the stopcock (E) and applying nitrogen pressure to the condensing chamber. While maintaining a positive pressure of dry nitrogen in the reaction chamber, the so3 capsule was introduced, frozen and broken with a glass stirring rod. Four hundred ml. of so2 were then collected into the condensing chamber and passed over into the main reaction flask and the so3 was dissolved and stirred magnetically. Another 100 ml. of so2 were then collected in the condensing flask and the hydrccarbon(s) 80 ·1ntroduced and the volume brought up to 135 ml. with s02• The cooling baths were removed and the solutions warmed to reflux temperatures. With the contents of both chambers " .~' : being stirred as rapidly as possible, the aJ;ene solution was transferred to the main reaction chamber. The time needed for combining the two solutions was usually 5 to 8 seconds. The resulting solution was stirred for about 10 minutes and then the apparatus was partially disassembled and the sulfur dioxi.de allowed to boil away. The temperature of the sulfonation reactions was measured,by means of a millivolt potentiometer. The reflux temperature was not exactly the same in all cases due to varying quantities of so3 .and hydrocarbon used in different experiments. The temperature for the sulfonation reactions was -12° to -13°c. After the evaporation of the sulfur dioxide, the reaction prdducts were dissolved in 200 ml. of ethyl ether which was then boiled off to remove the last traces of sulfur dioxide. Ether was added to make the solution back up to 200 ml. and this solution was extracted with four 50-ml. portions of distilled water. The collected aqueous extracts were then extracted twice with two 50-ml. portions of ethyl ether. The aqueous solution, containing the· sulfonic acid products of the sulfonation was then titrated with standardized NaOH to pH 7.0 following the titration wlth a Leeds and Northrup model 7401 pH meter. 81 For the isomer distribution experiments, aliquots of the titrated solution were measured into large quantities of inactive isomeric bromo~iihti(Mi~l:fflate salts for ..recrystallization. For the competitive' rate experiments, the . ~ titrated solution was evaporated to dryness and samples taken to determine the specific activity of the dx:ied s~lts. Samples of these dried salts_were also converted to the sulfonyl chloride for gas phase chromatographic analysis of the benzenesulfonic and bromobenzenesulfonic acid products.

Analysis of the Isomer Distribution Reactions Aliquot portions of the solutions of the, reaction products of the isomer distribution experiments, after titration with standar~ized Na.OH, were pipetted into three beakers--each containing about 40 g. of the pure, dry inactive salt of one of the isomeric bromobenzenesulfonic acid salts. To each beaker were added f-toluidine and concentrated hydrochloric acid in about 2-5 mole-pe~cent excess and enough hot water to dissolve the salts .. The p-toluidine salts of bromobenzenesulfonic acid crystallized without difficulty, the ortho salt formed large, w~ll- defined needles, the para gave much smaller and softer, almost cotton-like crystals, and meta formed thin flat plates. Samples of 0.4 to 0.5 g. were taken from each recrystallization for cou.r.,ting. Fewer slow recrystallizations were found to purify the isomers more 82 satisfact9rily than more faster and forced recrystallizations. Between 13 and 21 samples were normally taken of each isomer in each run. Abotit half of the colle~t.~d,samples were.dried overnight at 110°c under vacuum,'after which SO± 5 mg.- portions were weighed into counting vials and dissolved in l ml. water, 1 ml. absolute methanol and 15 ml.Kinard solution and counted in a Nuclear Chicago Unilux ambient temperature liquid scintillation counter. Colorization of f-toluidine salts of arylsulfonic acids noted in other work 14 was noted also to a slight degree in these experiments .but ·was not found to interfere significantly in the results.

Analysis of Competitive Sulfona~ion Reactions The _method developed previously for the determination of the products of benzene-chlorobenzene sulfonation was found to be fairly satisfactory for the benzene-bromobenzene system also. This method consisted of the sulfonation with non-radioactive s03, of a mixture of 1 benzene , labeled with c 4 and inactive bromobenzene by the same procedure used in the isomer distribution experiments except that the titrated salts were evaporated to .dryness and samples counted in Kinard solution as with the isomer distribution samples. The specific activity of the mixtures of benzenesulfonic acid and bromobenzenesulfonic acid were compared with the specific activity of the product of the 83 sulfonation of pure cl 4-labeled benzene (no bromobenzene) and the proportions of bromobenzenesulfonic acid in the mixtures being determined.as the amount of dilution of the radioactive benzenesulfotii~ a¢i~~aft. Another method of determination of the prq,p.uct distribution for the calculation of the relative rates of sulfonation consisted of the gas phase chromatographic separation of the benzeneSlllfonic acid and the bromobenzenesulfonic acid produced in the competitive reactions by . .,· converting the dried sodium salts.as obtained after the titration with standard NaOH~ to the sulfonyl chlorides by the action of phosphorus pentachloride,lOS which.were then . . injected into the gas chromatograph as a benzene solution. The sodium salts, o.s g., were weighed ;nto a reaction tube (4 in. female, 24/40 joint sealed at the end) and 3 g. phosphorous pentachloride were added. If the salts were completely dry, a drop of water was added before the PCl5 was weighed into the tube. The reaction tube was then placed in an oil bath maintained at 100-110°c and stirred mechanically for one hour by means of a long glass stirring rod inserted through a condenser mounted atop the reaction tube. The reaction mixture liquified as the sul~onyl chlorides were produced. Ten·ml. of benzene were _added through the condenser and stirring was continued for 5-10 minutes·,. The reaction tube was then removed from the o_il bath and allowed to cool as the excess PCls settled.· The benzene solution was then pipetted into a similar reaction 84 tube containing 5-10 ml. of water, which was then stoppered and shaken vigorously to hydrolyze the excess phosphorus pentachloride and the phosphorus oxychlor:J.de produced and suspended in the benzene layef • ., The benzene solution was then removed and dried over calcium chloride prior to i11jection into the gas chromatograph. Standard mixtures of sodium benzenesulfonate and sodium p-bromobenzenesul.fonate in ratios of about 50150 and 9515, the latter being appro.~i~4tely the ratio obtained in most.of the competitive reactions~ were prepared and •the sulfonyl chlorides produced in the manner described above. These were separated in the gas chromatograph an~ gave ' . ' results tha.t ~bowed that the· conversion to the sulfonyl cllloride was non-selective under the conditions used. The separation was ·carried out on a Varian model 200 Chromatograph through a four foot by 1/4 inch-stainless steel.·column packed with 4% methyl silicone gum r¥bber ·~ . . (SE..;30) on 80-100 mesh Chromosorb w, and detected by thermal .. conductivity. The column was maintained isothermally at 12s 0 c with a helium flow rate of 40 ml./min. Integration of the peaks was made by carefully cutting the peaks out and weighing on an analytical balance. An ~ffort was made to get the peaks as large as possible to reduce the e,rror inherent in the cutting-out process. 85 !>reparation of the Kj_nard Solution The Kinard6 8 solution was composed of five parts xylene (reagent, refluxed over~ight over sodium and distilled through a 30-plate helices packed column), five parts - p-dioxane (reagent,. di~tiiied and then filtered through a.column of activated alumina to remove peroxides), and three parts absolute ethanol (U. s. P.) in wJ:lich was dissolved 5 g./1. 2,5-diphenyloxazole (Packard PPO, ..... scintillation grade) and 50 mg./1. a-naphthylphenyl- oxazole (Packard a - NPO, scintillation grade) and 80 g ./1. reagent grade naphthalene. The Kinard solution was.stored in a brown bottle in the refrigerator and used as fresh as possible since: peroxide formation causes deterioration. LIST OF REFERENCES LIST OF REFERENCES

1. L. J. Andrews and R. M. Keefer, J. Am. Chem. soc., 1.l, 3113 (1950).

2. H. E. Armstrong, J. Chem. Soc~, 51, 258 {1887). From Ref. 110.

3. E. Bamberger and J. Kunz, Ber. 30, 2274 (1897). From Ref. 110.

4. R. Behrend and M. Merte.,lmann, ~, 378, 352 (1911). From Ref. 110. s. u., Berglund-Larsson and L. Melander, Arkiv Kemi, 6, 219 (1953). - 6. M. L. Bird and D. K. Ingold, J. Chem. Soc.,~, 918. 7. T. G. Bonner, F. Bowyer and.G. Williams, !_bid-!., !Jl53, 2650. a. J. I<. Bosscher and H. Cerfontain, ibid., 1968 B, 1524. 9. J. K. Bosscher and H. Cerfontain, Tetrahedron, 24, 6543 (1968). 10,. E. A. Brolvn, Doctoral Dissertation, Brigham Young University, 1967, p. 17-22. 11. Ibid., P• 32. 12. Ibid., P• 35. 13·. Ibid., P• 67-72. 14. Ibid., P• 79-81.

15. Ibid., P• 101. 16. H. c. Brown, J. Am. Chem. soc., 74, 191 (1952). 17. H. c. Brown and J. D. Bradf, ibid., ~.3570 (1952). 18. H. c. Brown and G. Goldman, ibid., 84, 1650 (1962).

87, 88 19. H. c. Brown and G. Marino, ibid., 84, 1658 (1962). 20. H. c. Brown and K. L. Nelson; ibid., 75, 6292 (1953). 21. H. c. Brown and A. Neyens, aJ?!d,, 84, 1655 (l 962). 22. H. c. Brown and Y. Okamoto, ibid., §Q, 4979' (1958). 23. H. c. -Brown and L. M. Stock, ibid., 79, 1421 (1957). 24. H. c. Brown and R. A. Wirkala, ibid., 88, 1456 (1966). 25. s. Y. Caille and R. J. P. Corriu, Chem. Comm.,~, 1251.

26. H. cerfontain, H.J. Hofman and A. Telder, Rec. Trav. Chim., ~, 493 (1964). ·. -

27. H. cerfontain, F. L. Sixma and L. Vollbracht, ibid., .§1, 659 (1963). 28. H. Cerfontain and A. Telder, Proc. Chem. soc., illi, 14~ 29. H. Cerfontain and A. Telder, Rec. Trav. Chim., 84, 545 (1965).

30. H. cerfontain, A. Telder and L. Vollbracht, ibid., ~, 1103 (1964).

31. Ibid., 84, 1613 (1965).

32. E. Chambers and G. w. Watt; J. Org. Cheu\!., 6, 376 (1941)~ 33. N.- i>. Cheronis and J.B. Entrikin, "Semimicro Quali- tative Organic Analysis," 2d ed., Interscience Publishers, Inc., New York, N. Y., 1957, p. 696. 34. N. H. Christensen, Acta Chem. Scana., 15, 1507 (l9Eil). 35. Ibid., 17, 2253 (1963).

36. Ibid.,!!!, 954 (1964). 37. Ibid., lQ., 1955 (1966). 38. Ibid.,£!,, 899 (1967).

39. F. E. Condon, J. Am. Chem. Soc., 71, 3544 (1949). 40. w. s. Cowdrey and D. s. Davies, J. Chern. Soc., !2,i2, 1871. 41. M. J. s. Dewar, ibid., lli.§., 406, 777. 89 42. M. J. s. Dewar, "Electronic Theory of organic Chemistry," Clarendon Press, oxford, 1949, p. 168. 43. E. Dresel and c. N. Hinshelwood, J. Chem. Soc., ill!, 649. 44. c. Eaborn, ibid.,~, 4858. 45. c. Eaborn and J. A. Waters, ibid., !2fil-_, 542. 46. c. Eaborn and D. E. Webster, ibid., 1957, 4449. 47. Ibid., .!2.§.2., 179.

48. D. D. Eley and P. J. King, Trans. Farada:f: Soc., 47, 1287 (1951).

49. L. N. Ferguson, A. Y. Garner and J. L. Mack, J. Am. Chem. soc., 76, 1250 (1954). SO. E. E. Gilbert, .. Sulfonati.on of Aromatic Hydrocarbons," in "Chemistry of the Petroleum Hydrocarbons, .. B. ·.T. Brooks, c. E. Boord, s. s. Kurtz !;:lnd L. Schmer- ling, ed., Reinhold Publishing Corp., New York, N. Y., 1955, Vol. III, chap. 58, P• 601. 51. w. E. Gilbert, "Sulfonation and Related Reactions," Interscience Publishers, New York, N. Y., 1965, Chap. 1. 52. ,E. E. Gilbert and B. Veldhuis, Ind. Eng. Chem._,, il, 2300 (1955). .. -

53. R •.' J. Gillespie and E. A. Robinson, Can. J. Chem., 39, 2189 (1961). 54. v. Gold and B. W, v. Hawes, J. Chem. soc., l951, 2102. 55. V. Gold and D. P. N. Satchell, ibid., ill&, 1635. 56. E. s. Gould, "Mechanism and structure in organic Chemistry," Holt, Rinehart and Winston, New York, N.~ Y.; 1959, p. 446. 57. L. P. Hammett, "Physical Organic Chemistry, .. McGraw-Hill Book Co., New York, N. Y., 1940, P• 309. 58. J. Hine, "Physical organic Chemistry," McGraw-Hill Book co •• New York, N. Y., 1962, p. 72. 59. Ibid., P• 73. 60. Ibid., P• 348. 90 61. I~id.,·p. 353.

62. A. E. Holleman and P. Caland,. ~, 1,i, 2504 (1937). 63. G. Illuminati and B. M.liirid,.J. h.'ll. Chem. Soc., 1!!, 4975 64. c. K. Ingold, A. Lapworth, E. Rothstein and D. Ward, J. Chem. Soc.,~, 1959. 65. c. K. Ingold and F. R. Shaw, .!J?_;_d., ill.I, 2918. 66. o. I. I

68. F. E. Kinard, Rev. Sci. Instr., 28, 293 (l957j. 69. w. Klatt, z. anorg. Chem., lli, 189 (1937). From Ref. 86. 70. H. s. Kl.ein and A. Streitwieser, Jr., Chem. and Ind., ill.!, 180. 71. A. Klit, and A. Langseth, ~- phY.SiJc. Chem., ill., 65 (1936). 72. J. Knight, Master's Thesis, Brigham Yotmg University, 1957, p. 29. .

73. c. w. F. Kort and H. Cerfontain, ~ Trav. Chim., !!§, 865 (1967).

74. H. G. Kuivila and A. R. Hendrickson, J. Am. Chem. Soc., H, soGa (1952). 75. A. Liberles, "Introduction to Molecular Orbital Theory," Holt, Rinehart and Winston, Inc., New York, N. Y., 1966, Chap. 3.

76. D. A. McCaulay and A. P. Lien, J • Am. Chern. Soc., l,3, 2013 (1951).

77. c. w. McGary, Jr., Y. Okamoto and H. c. Brown, ibid., §1., 3037 (1955). 78. H. Meerwein, G. Dittmar, R. Goellner, K. Hafner,F.Mensch and o. Steinfurt, Ber., 90, 841 (1957),. c. A., 52, 9003b (1958). - - . -

79. L. Melander, Nature, 1§1, 599 (1949). ao. L. ~elander, Acta Chem. Scand., 3, 95 (1949). 91 81. L. Melander, Arkiv Kemi, 2, 211 (1950). 82. G. T. Moody, Chem. News, §1, 34 (1893). From Ref. 110. 83. Y. Muramoto, Kagaku to K99yo Osaka, 33, 259 (1959). From Ref. 110. ·

84. P. c. Myhre, M. Beug am.'I..L •.. 1.. James, ~Am. Chem. Soc., .2.Q., 2105 (1968).

85. K. V. Nahabodian and H. G. Kuivila, ibid., 83, 2167 (1961). - 86. K. L. Nelson and H. c. Brown, "Aromatic Substitution, Theory and Mechanism," in "Chemistry of the Petroleum Hydrocarbons," B. T. Brooks, c. E. Boord, s. s. Kurtz. and L. Schrnerling, ed.~ Reinhold Publishing Co., New York, N. Y., 1955, Vol. III, Chap. 56, P• 491. 87. Ibid., p. 503.

88. Ibid., p. SOS. 89. Ibid., p. 539. 90. K. L. Nelson, ttsulfonation, .. in "Friedel-Crafts and Related Reactions," G. A. Olah, ed., John Wiley and Sons, Inc., New York, N. Y., 1964, Vol. III, Pt. 2, Chap. 42. . 91. A. Nilsson, Acta Chem. Scand., 1!_, 2423 (1967).

92. E. Noelting, ~. 9, 1797 (1876). From Ref. 86. 93. R. o. c. Norman and R. Taylor, "Electrophilic Substitution in Benzenoid Compounds, Elsevier Publishing co., New York, N. Y., 1965. ' 94. G. A. Olah, ed., "Friedel Crafts and Related Reactions," John Wiley and Sons, Inc., New York, N. Y., 1963, . Vol. I.

95. G. A. Olah, s. H .. Flood, s. J. Kuhn, M. E. Moffat and N. A. Overchurch, J. Am. Chem. Soc., ~, 1046 (1964 ). 96. G. A. Olah, s. H. Flood and M. E. Moffatt, ibid., 1065 (1964).

97. G. A. Olah and s. J. Kuhn, j._Eid., 80, 6541 (1958). 98. Ibid. , .§.!, 3684 (1962). 92 99. G. A. Olah, s. J. Kuhn ands. H .. Flood, ibid., §1., 4571 (1961 ). 100. Ibid., 4581 (1961). 101. Ibid., 84, 1695 (1962) • . - 102. G. A. Olah, s. J, Kuhn, s. He Flood and B. ·A. Hardie, J. Am. Chern. Soc .. ,,86,.1044 (1964). 103. G. A. Olah, s. J. Kuhn and B. A. Hardie, ibid., 1055 (1964). 104. G. A. Olah and N. A. overchurch, ibid., 87, 5786 (1965). 105. J. s. Parsons, J. Gas. Chrom, 5, 254 (1967). 106. J. s. Reese, J. Am. Chem. Soc., 54., 2009 (1932) •. 107. J. c. Robertson, Doctoral D~ssertation, Brigham Young University, 1962, ~• 82 •. 108. Ibid., PP• 92, 93 and 95 .• 109. w. H. c. Ruggeberg, T •. w. Sauls ands. L. Norwood, J. Org. Chem., 20, 455 (1955). 110. w. J. Spillane and F. L. Scott, Tetr~edr~, 24, 5011 (1968). 111. L. M. Stock and F,. w. Baker, .J. Am. Chem. Soc., 84, 1661 (1962). ,. -- -

112. E. M. Thompson, Doctoral Dissertation, Brigham Young University, 1963, P• 42. 113. J~, p. 44. 114. o. R. Vicary and c. N. Hinshelwood, J. Chem. Soc., !ill, 1372. 115. K. D. Wadsworth and c. N~ Hinshelwood, ibid.,~. 469. 116. A. c. M. Wanders and H. Cerfontain, Rec. Trav. Chim., 86, 1199 (1967). 117. A. c. M. Wanders, H. cerfontain and c. w. F. Kort, ibid., 301 (1967). APPENDIXES APPENDIX A

GENERATIONOF DATATO TEST THE EFFECTS OF SECONDARYREACTIONS GENERATION OF DATA10 TEST THE EFFECTS OF SECONDARYREACTIONS

That under t.he conditions of secondary ~eactions ....._ which are less selective than the primary reactions; the final product distribution consists of a proportionately larger amount of the product of the less reactive substrate, was simply demonstrated by Brown by the use of inequalities. If BSl > CSl and BS2 > CS2 BSl BS2 and csl > CS2

then CSl BSl csl BS2 --·-BS2 X CSl > BS:2 X CS2

and BSl csl BS2 > CS2 or BS2 CS2 BSl < CSl which shows that CS2 is produced in larger quantities relative to CSl than is BS2 to BSl. Brown concluded that under these conditions, secondary reactions would account for the variations observed in the product distributions of competitive sulfonations of chlorobenzene and benzene. It seemed

95 96 desirable to go beyond the derivation based on inequalities and generate data, actual numbers, to test the influence of secondary reactions. For the purpose of generating artificial data, t~10 mechanisms were considered. The first, suggested by current reports in the 11.terature, involved, as a primary product, the respective pyrosulfonic acids

AH + 2 S ---➔ ASSH

BH + 2 S -----➔ BSSH which then, in some manner, sulfonated more of the starting

..., aromatic reactants with a different selectivity than the primary reaction.

ASSH + AH ASSH + BH ASH + BSH BSSH + AH ASH + BSH BSSH + BH ---➔ 2 BSH

It was assumed for the purposes of the calculat:i.on,

(l) tha·t the reaction rate ratio for the primary reaction was kA/kB = 20 and for the secondary reaction k,A/kB = 2, (2) that the pyrosulfonic acids were equally selective in the.secondary reaction, that, since the secondary reaction is reported in the literature to -be slower than the primary reaction, (3) the secondary reaction was half as fast as the primary reaction, 1. e. half of the pyrosulfonic acids in the reaction mixture during ~...nyreaction period were assumed to have participated in the secondary reactton. 97 The mechanics of the calculation were carried out in

the following manner. The px·oduct distribution of the pyrosulfonic acids produted.1H tH@~rimary reaction was calculated using the Ingold equation

log (A1/Af) log (~'Bf)

where the subscripts refer to initial ~nd final concen-

trations. A1, B1 ~"ld ~ were cho.sen as convenient, for instance, Ai= Bi= 20 and Af = 19, then Bf was calculated from

The differences between Ai and Af and between B1 and Bf were .the quantities of primary products (pyrosulfonic acids) produced during the initial reaction period. The average concentration of . pyrosulfonic ac·ids (not considering the secondary reactions) would have been (Ai - Af) + 2 Bi - Bf ) ( 2 - • In the slower (half as fast) secondary reactions, one-half of the average concentration of the ASSH and BSSH was consumed in. the secondary reactio~s. Since the primary products, ASSH and BSSH, were considered equally reactive, one-half of the ASSH and BSSH participat~d in the secondary reaction. For the secondary reaction, kA/kB = 2, so that one-third of the aromatics sulfonated in the 98 secondary reaction was BH and. two-thirds AH. The total ASH produced in the initial reaction period is derived from the ASSH,· (Ai~ Af ), and from the

AH sulfonated in the secondary reaction,

2/3 ((Ai - Af) + (B1 - Bf) ) 4 •

The .BSH produced is similarly calculated,

The quar,tities of the various pro~ucts after a second reaction period are calculated in the same manner. The new Ai was taken as Af from the first reaction period less the AH consumed in the, seconda.ry reaction during that period. The Bi for the second period was similarly determined and a new Af chosen and a n~w Bf calculated from the Ingold equ~tion. In the second and subsequent reaction periods, the ASSH (and BSSH) involved in the secondary reactions was one-half the total ASSH (and BSSH) in the reaction mixture during the period, i.e. one-half of the average amount produced in the period plus one-half of that remaining from previous reaction periods. From the second reaction period, the ASH produced was given by

+ 99 and BSH was

+ 1/3 ( (A~;? - Af2) 4 .

The subsequent calculations ~ere made in the same manner, each time utilizing the newest.Af and Bf for the following

Ai and B1e The calculations for the case where the primary reaction was less selective than the secondary reaction

(the. rate ratio of the primary r,action k,Af'kB = 2 and of the secondary reaction kykB = 2cf) ~ere made in. a completely analogous manner.

The second mechanism considered in this study · ~ involved the production of the su~fonic acids in the primary reaction, but part of these products reacted further with the electrophile to give the pyre>sulfonic acids which then

AH + S ---=+- ASH ___ _. BSH BH + S sulfonated more of the starting' .aromatic compounds with

ASH + s ASSH BSH + . s BSSH a different selectivity than the initial (primary) reaction.

ASSH + AH 2 ASH ASSH + BH ---+ ASH + BSH BSSH + AH ASH + BSH BSSH + BH 2 BSH 100 As the concentrations of ·the ftnal products increased, larger proportions were produc~d through the secondary (less selective) reactions. The conditions and a.ssq.dtp'b.idns for this mechanism

~ . included, (1) kp/kB = 20 for ~he primary reaction and k~B= 2 for the secondary reaction, (2) one-fourth of the total final products, ASH and BS_H, entered into the secondary reactions in each reaction period, (3) ASH and BSH were equally reactive toward the electrophi.le, (4) the pyrosulfonic acids were equally reactive in sulfonating starting aromatic compounds, and (5) it was assumed that the pyrosulfonic acids reacted rapidly with the aromatic compounds so that the concentrations of the pyrosulfonic acids remained negligible throu.ghout the reaction. The mechanism of the calculation involved determining the pr~duCt distribution of the primary reaction as with the first,mechanism by means of the Ingold equation. The products of the primary reaction were obtained from

ASH = (Ai - Af)

BSH = (Bi - Bf) one-fourth of the average concentration of the primary products during this reaction period reacted with more electrophile so that the ASSH and BSSH produced was

ASSH = (Ai 8 Af) , and BSSli = (Bi ; Bf) • The amounts of AH and BH reacted in the secondary r~actions were, for AH, 101

2/J ( (~\i - Af),: (Bi - Bf) )

and for BH, l/3((Ai - Af) + (Bi, - Bf) ). The total ASH and ' ·, 8

BSH produced in the initial r~ad~icrl.,.~ . period was given by

Af) ASH= 8 ·+··AH ( consumed in the secondary ; · reaction) (Bi - Bf) BSH = + BH(consurued in the secondary 8 reaction) ,

For· the next Z'eaction ~eriod, A was Af less the 12 1 amount of AH used in the secoodary reactions in the first reaction period. B12 was de,termined by the same relation- ;, •· ship with Bf• A new Af was chosen and a new Bt was 2 2 calculated fran the Ingold eqL1ation. APPENDIX B

DESIGN OF A FLOWAPPARATUS -FOR THE STUDY OF

THE BROMINJl..TION' OF OLEFINS · DESIGN OF A FLOWAPPARATUS FOR THE STUDYOF THE BROMINATIONOF OLEFIN$

The results of a kinetic experiment are only as good as the methods involved in conducting the investigation. The critical factors are (l) the accuracy of the analytical. techniques, (2) the precision of the method for determining the reaction time, and (3), especially for rapid reactions, the efficiency of mixing the reactants. The analytical methods used by Gurney and Duvall in previous studies of the brornination of olefins in this laboratory were standard proven techniques involving the titration by thiosulfate of .iodine produced by th~ reaction ofa KI quench solution with excess bromine in the reaction mixtu.re. It was hoped that the present work would lead to a method offering improved mixing and timing procedures for the reactions For rapid reactions, the mixing process must be very fast and efficient. It is necessary that the half time of the reaction be long compared to the time required for mixing the reactants. A large variety of mixing chambers has been designed for continuous flow systems used in kinetic investigations. Some utilized several jets arranged tangentially about the mixing:chamber which gave about 97%

103 104 mixing in 0.0003 to 0.0004 seconds. Simpler mixers, even capillary bore three-way stopcocks, have been used for reactions having half times greater than 0.01 seconds. Mixing rates have a high thermal coeff :!.cient, · being of the • "! '· ., '•·, ., y order. of three times faster ·a~ 40°c than at 20°cc The coo.tact time of the reactants in a flow system depends upon the' flow rate of the reactants and the volume of the reaction tube by the equation

Contact time = Volume of the reaction tube Flow rate

Some analytical methods, such as spectrometry,, may be applied at various positions along the reaction tube (if it is constructed out of the proper materials); ho~ver,, where the analysis of the reactants c1nd/or products is ?Y chemical methods,, the reaction must be.quickly and completely quenched after a given contact time. The flow rate may easily be determined by means of a calibrated ~apillary flowmeter placed between the nitrogen source and the bromination apparatus. The flow rate is related to the pressure differential across the capillary tube as indicated by the height of the liquid in the internal manometer tube. Calibration of the flowmeter involves measuring the height of the manometer liquid by means of a cathetometer while determining the flow rate with a soap film flowmeter. The calibration of the capillary flowmeter is showa1 bel~w using di-n_-butyl phthalate as the manometer 105 liquid. The range of the flow meter may be increased by

using a different capillary or, ,bY using a manometer liquid : ' with a density different than di-n-butyl phthalate and !£., ,:; '/,:i:•viscosities•of the solutions involved. The roixing ratio was determined t·or dichloromethane by - 130

l.201- .. 110 ,,. ...,~: 0 · .100

90 • 0 ....C ~ 80 • 0 ~ 70 0) ~-- 60 f-1 0 ,.

~- 501- ....,t fx,· 401- 0

30 0 20 0 lOt-

0 5 10 15 20 25· 30 35 '40 45 · 50 55 60 65 70 75 80 85 90 95 100 Flowmeter Readi~g mm. n. butyl phthallate i""' 0 Fig. 21.--calibration of the flowmeter. 0\ 107 determining the relative dilution of a 0.1 N iodine solution in dichloromethane with pure dichlor.omethane as the dichloromethane and iodine-

An apparatus, D), and the reactants are mixed at E. initial.• portion of the reaction mixture ,is discarded. After the system has reached dynamic stabili'ty, flasks containing a known amount of KI solution, which is being vigorously stirred throughout the process, and has been preweighed, are placed so as to receive the reaction mixture from the 109

, Fig. 22.--The bromination apparatus. 110 reaction tube, F .. A series of six to eig-ht flasks may be prepared and successively· used to give six to eight determinations ' for a given set of conditions fran each full charge of the apparatus. The flowmeter is read during the process of collecting these :s&nples of the reaction mixture. It will not be convenient to stop the react 0:i.on before the reservoirs are emptied since a backflush of the +eaction tt1be will ··contaninate the starting reactant solutions. After the solutions a.re quenched, the' flasks ?1re reweighed to determine the amounts of the reaction mixture received in each. From the mixing ratio for the conditions ,. of the reaction, the initial,concentrations of the reactants may be calculated. The final concentrations are determined by the standard analytical procedures previously developed. The contact time is determined by the volume of the reaction tube used and the flow rate. APPENDIX C

A RESEARCHPROPOSALa A S"rUDYOF THE STABILITY ANO REACTIONS OF THE CYCLOPENTENYLFREE RADICAL _, A Research Proposal in Organic

.' Chemistry in Partial FUlf illment of , ! the Requirements .·for the Degree of Doctor of Philosophy

A STUDYOF THE STABILITY AND REACTIONS OF THE CYCLOPENTENYLFREE RADICAL

June 19, 1968 laOO P.M. 242 ESC

by Sullivan E. Blau A STUDY OF THE S'r'ASILITY, .. . . •'•,,. Ade ,, REACTIONS OF THE CYC.LOPENTENYLFREE RADICAL

The cyclopentenyl free'radical has been proposea 1 , 2 as an intermediate leading to various secondary products in the decomposition of the cyclopentyl free radical. Wh_ile it has been produced by irradiation of cyclopentane 3 and cyclopentene 4 at low temperatures and ESR spectra·obtained while in suspension in a solid matrix, 4 the cyclopentenyl radical has not been shown without ambiguity to exist in systems analogous to Gordon's at higher temperatures. 516

Objective of tJle Proposed Stugy_ ' The objective of this study will be to produce the cyclopen°t'enyl free radical in an unambiguous reaction, determine its thermal stability., i.e. the lowest temperature at which ring opening and/or fragmentation occurs, and propose a mechanism for its decomposition based upon the products derived from its decomposition. The cyclopentenyl free radical is expected to be resonance stabilized and, thus, to have chemical propert.ies analogous to the allyl free radical. The reactions expected from such radicals will be hydrogen abstraction and recombination which are the only modes of decomposition of the allyl free radical. 2 Other modes of decomposition are

113 114 ~vailable to the·cyclopentenyl radical; products may include the cyclopentadienyl free radical and hydrogen, fran the elimination of molecular hydroge~ 7 fran the cyclopentenyl radical and pentadienyl radicals resulting fran ring opening. In the decomposition of a number of alkyl radicals a common mode of deccmposition leads to an allyl radical and an ~lefin. 8· If an analogous reaction occurs in this system, acetylene would be produced along with the allyl radical. Each of the different kinds of radicals produced would be expected 'to react by hydrogen abstraction or recombination to give corresponding products. Toluene will be used as a solvent to provide a substrate for hydrogen abstractions and recombination.s while being but· slightly susceptible to secondary decomposition. It is expected that sign~ficant. information as to the mechanism of decomposition, such as direction of ring opening and which bonds undergo unpairing of electrons in fragmentation processes, would be obtained from the relative radioactivities of the final products of the decomposition of c14-labeled cyclopentenyl free radicals.

Proposed Experimental Approach

Cyclopentenyl free radicals will be produced by the thermal decomposition of azocyclopentene-2. This compound has not been previously prepared but it should be readily S~'"llthesized by adaptations of known reactions. 9,lO,ll 115

The experiment will be carried out by injecting azocyclopentene-2 in tolu~ne solution into a reaction tube maintained.· at constant .temperature in a combustion tube furnace. The reaction tuba "ill b.e integrated into (and .isolated from) the gas flow $YStem of a gas chromatograph by means of a six-way gas sampling valve. This valve will° also measure samples of reproducible volume of decomp~sition products into the gas chromatograph~ This will allow an approximation of the kinetic~ of the reaction if deco~position of the azo compound occurs at a t~mperature low enough that the reactions of. the f;-ee radical products are slow.· The products of the react~on will be separated in the gas r..hromatograph, collected and characterized by infrared. and nuclear magnetic. resonance methods.~ Azocyclopentene-2-1-cl4 will be synthesized and decomposed in the normal manner arld the products will be collected and the relative radioacti.vities deterntined by standard counting tec."'iniques. 116

BIBLIOGRAPHY.· al - ' (Jo

l. A. s. Gordon, can. J. Chem., il, 57CJ (1965). 2. J. R. McNesby and A. s. Gordon, J. Am. Chem. Soc., .12, 4593 (1957).

3. R. A. Holroyd and G. W,. Klein, J. Am. Chem. Soc.,.§!, 2922 (1962). 4. I. Nitta, T~ omae, s. Ohnishi, K. Kuwata and a. Sakurai, . Nippon Hcshasen Kibunshi Kenkyu Kyokai Nempo., .§, 295 (1964); C. A. §.2, 5337 '(1966) • s. H. E. Gunning and R. L •. Sto~, can. J. Chem., 42, 357 (1964) • .· 6. T. F. Palmer and R. p,. Lossing, can. J. Chem., _,43 565 (1965) • 7. A. s. Gordon arid s .. R. Smith, !t.! Am. Chem. soc., _§£r, 331 (1960).

a. A. Fi~h; ~t. Rev., .!§., 243 (1964).

9. A. Mur·ray, and D. L. Williams, ed., 11orgs.nic Synthesis with Isotopes, 11 Pt. l, Intersci~.nc~ Publishers, Inc., New York, 1958, p. 659 • .. 10. A.H. Blatt, ed., "Organic• Syntheses," Coll. Vol. 2, John Wiley & sons, Inc., New York, 1943, p •. 116. 11. E. c. Horning, ed., "Organic Syntheses," Coll. Vol. 3, John Wiley & Sons, Inc~, New Yorkr 1955, p. 350. 12. N. Rabjohn, ed., "Organic syntheses," Coll. Vol .. !, John Wiley & Sons, Inc., New York, 1963, p. 238. APPENDIX D

MANUSCRIPT FOR PUBLIC:~TION: REACTIV!;TYAND SELEC'TIVITY OF TBE REAC';T.'IONOF SULFUR

< TRIOXIDE AND:BROMOBENZENE REACTIVITY AND SELECTIVI'l'Y OF THE REACTIONOF

SULFUR TRIOXIDE AND BROMOBENZENE

K. Le Roi Nelson and Sullivan E. Blau 1 Department of Chemistry Brigham Young University Provo, Utah 84601

(1) From the Doctoral Dissertation of s. E. Blau, Brigham Young University, Provo, Utah, 1969.

ABSTRACT

The sulfonation of bro~obenzene with sulfur trioxide in liquid sulfur dioxide at -12 to -13°c was studied to determine its relative reactivity. The characteristics of the data suggest the

occurrence of secondary reactions at high s03 concentrations and low benzene/bromobenzene r~tios.

The relative rate constant ratio, k¢ 8/k¢Br' was found_to be 28.6 in a competitive reaction with

benzene at low S03 concentrations and high benzene/ bromobenzene ratios. The isomer distributio~ was

found to be less influenced by the secondary reactions, and the values obtained under the conditions of thi.s f.itudy were o.15 ± 0.02% ~,

0.91 :t: 0.04% ortho and 98.93 ± o.14% ~r~. The partial rate factors as calculated from these values are mf = 0.00018, Qf = 0.0096 and Pf= 0.208~ 118 119

Aromatic sulfonatio11s, especially when so3 is used '.'; as the ·sulfonating agent, 'have ·several characteristics which .· are unique among electrophil~c aromatic substitution reactions . • They. are re\ritt!'i1'1~.•·bhtlie't relatively mild conditions, 2 they exhibit:a kin~tlc isotope effect, 3 the

(2) E. Ba.niberger and J. Kunz, Ber., 30, 2274 (1897). R. Behrend and M. Mertelmaiin;· Anri;, 378, 352 (1911.) .. A~ F. Holleman and P. Cqland, Ber., 44, 2504 (1937). O. l,.· Kachurin and A. A.,· Spryskov, Izv. V. Sh. Khim. lChim. Tekhnol.,, 1958, 52. G. T. Moody, Chem. Ne'WS:- §J_,, 34 (1893). ¥:'""Muramoto, !_

(4) (a) D.R. Vicary and c. N. Hinshelwood, J. Chem. Soc., 1939,.- 1372. (b) K. D. Wadsworth and C. N .. Hinshelwood, "Ibid., -·1944, 469. (c) E. Dresel and c. N'. Hinshelwood, 'Ibic:l., 649. (d) J. K. Boo~cher and H. Cerfontain, Tetrahedron, 24, 6543 ( 1968) • · they are complicated by secondary reactions, 4d,S they are

(5) (a) J. K. Bosscher and H•. Cerfontain, J. Chem. Soc., (B) 1968, 1524. (b) E. A. Brown, Doctoral Dissertation, Brigham Young University, Provo, Utah, 1967~ (c) H. ··Cer'fontain, A. Telder and L. Vollhracht, ~ec. Trav. Chim., 83, 1103 (1964). (d.) N. H. Christensen; Acta Chem. Scand~, 15, 1507 (1961), ibid., 17, 2253 (1963), ~., 1s-;-9s4(1964). -- - -·,------120 also found generally to give a poor fit to the reactivity- selectivity relationship 6 predicted by equation (1)7 with

(6) H. c. Brown and K.- L. Melson, J. Am. Chem. Soc., ll, 6292 (1953). ,\.,· I •:f!- ./ (7) c. w. McGary, Y. Okamoto and H. c. Brown, lli_g., 77, 3037. (1955).

8 the values of crP + and crm. + as tabulated for the various

(8) H. c. Brown and Y. Okamoto, !B!9., ~, 4979 (1958). monosubstituted benzenes. It is not improbable that all of these imusual characteristic·s are rela.ted to,. some particular peculiarity in the reaction mechanism.

a + log Pf = (----P----) log ~ (l) fJp+ - Om+ . Inf

Hinshel'l'flO~d4a, 4b, 4c hoted that in addition to the second order dependence on sulfur trioxide, there was~ "retardation by the product" of the .reaction, This corre- lates well with the observation of Booscher and Cerfontain 4d that a rapid initial reaction used up half the aromatic compound which w~s equimolar with the so3 at the start of the reaction, after which the aryl sulfaiic acid was produced in a slower subsequent step. Christensen 5d detected the formation of a ltl iodobenzenesulfonic acid-sulfur trioxide species. The formation of pyrosulfo~ic acids is generally conceded; .however, they are usually ignored in kinetic 121 studies, being considered as transient species that store

and then release so3 or otherwise' participate in furthe1· sulfonation steps. Christensen 5dcconsiders the pyrosulfonic acids as active sulfonating ~geats being only a little less active than s03. He also isolated sulfonic acid anhydrides and studied their reactivity~ He concluded that they, too, are sulfonating agents, being a little less active as such

than so3 and the pyrosulfonic acids. :.Booscher and Cerfontain 4d also observed the formation of the sulfonic acid-anhydrides but suggested that their influence in t..lte reaction is limi t.ed especially when the reaction time is kept short. sv,lfones are also common '.products in So'3 sulfonations. Sa, 9 ,lo They are,. however: produced in very

' (9) E. E. Gilbert, "Sulfonation·and Related .Reactions," I~terscience Publishers, New York, N. Y., 1965, p. 66. (10) · w. H.· c. Ruggeberg, T. w. Sauls and s .. L. Norwood, ,, J. org. Chem., 20, 455 ('1955). ·• limited quantities and therefore exert negligible influence on the kinetic studies of most so3 sulfonation reactions~ Bosscher and Cerfontain 4d suggest that the removal of the hydrogen from the rr-complex is intramolecular. '!'he function of the second S03 mol~cule is almost certainly not that of a typical base sinoe the.solvent is more basic than · 11 S03• .

(11) J. Hine, "Physical organic Qhemistry," McGraw-Hill Book New Yorlc, N. 19E2, P• 353. co.,· y;, ,, 122 While a large majority of electrophilic reagents undergo aromatic substitutions on toluene such that within experimental err9r, the plot of log Pf vs. log Pt/mf falls on a s~x:aight line, the !4ul~bnat1.6ti bf toluene gives deviating· results. 5h,Sc Similar 'deviations have also been -~ noted_for chlorobenzene, 5b i:odobenzene 12 and in this work

(12). E. M. Thompson, Docto~al Dissertation, Brigham Young University, Provo, Utah, 1963. on bromobenzene.

Results and Discussion

The sulfonation of bromobenzene with so3 in liquid S02 at a reflux:.tng temperature between -12 and -13°c

(depending on tho concentrations ·-of the reactants) has been studie~ to determine the partial rate factors for the reaction and to measure its relative reactivity under these .. conditions. The methods used in this study, for: t.he most part, h~ve been previously deve{oped and tested'..Sb,l 2 The reaction rate constant ratio was determined by sulfonating mixtures of ben~ene and bromobenzene, analyzing the products to determine the amounts of benzenesulfonic acid and bromobenzenesulfonic acid resulting from the reaction'and applying t])e Ingold equation 13 .(2) •

(13) c. K. Ingold,. et al., J. Chem. Soc., ~..;.• •.1959, ibid., --1927, 2918. ., 123

log (A1/Af) (2) log(B1/Bf) where A1 and Bi afe the tn!ti,al "ca.b¢entra.t1ons of the reactants.and Af and Bf are the final concentrations. The ., Ingold equation is genfl..rally considered to be applicable to reactions in which the reaction is first order with respect to A and B although independent of the order of the electrophile as long as its kinetic order is the same for the reaction with both A and B. 14 The usual test for the

(14) M. ,.J. s. Dewar, T. Mole and E. w. T. Warford, J. Chem. Soc., 1965, 3576. ---,-·------applicability of the Ingold equation is the independence of

.' kA/ka from the initial ratio of A and B. That this criterion is not always val'id has been pointed out ..5 b,i5

(15) K. L. N~lson and E. A. Brown, a paper presented at the national meeting of the American Chemical society, .San Francisco, April, 1968.

Benzene and branobenzene were competitively. sulfonated and the reaction products were worked up as described in th~ experimentalsection. The total moles of sulfon:i.c acid products were.determined by titrating with standardized NaOH. The sulfonic acid salts were ~vaporated to dryness and samples were converted to the sulfonyl" chlorides and injected in a benzene solution into a gas chromatograph. Relative amount::>sof the two sulfonyl, chlorides·were determined by- int~grating- the peaks. 124

Table l shows the results of twenty competitive reactions. It is obvious that the relative rates are not constant, and the differences are apparently due to more ... '). ;' ·.· .,,;,. ,~- ...';.;. :l:' .J, than variations in the st:'arfing~ rat.ib of the aromatic reactants. The total moles of the reactants were maintained essentially constant between runs. The differences may be

derived. . fran':;. mechanical functions such as variations in stirring rates, etc., or from diff'erences in the basic reactiono While a dependence on the so3 concentration is not obvious, plots of the ratio of the sulfonic products against the S03 concentration gives a "family" of chara6teristic curves (Figure 1) for the various initial ratios of benzene and bromobenzene. There is considerable scatter in the points where the•initial ratio of benzene and bromohenzene is l.O. The signiflcance cf this is not obvious. That this scatter is significant is pointed up by the consistency of the points at other ratios where the experimental error is expected to be greaterD

It is readily seen that at high initial concentrations of S03, the ratio of the products is rnuc.~ less a function of the initial ratio of the aromatic reactants than at low s0_3 concentrations. The indication of the various curves is that at some sufficiently high S03 concentration, the ratio of sulfonic acid products may become linear, 125

independent of so3 concentration and, perhaps, even independent of the initial ratio of aromatic reactants. The final result of the reaction is shown by this plot to be an increase in t.G~ptbutact.ic,n of bromobenzenesulfonic acid relative to benzenesulfonic acid as the initial ratio of benzene and bromobenzene is reduced and as ~( the &"tlount of so3 involved in the reaction _is increased. · Such effects might be expected under certain conditions, as, for instance, if the kinetic ord~r of benzene were considerably higher in the reaction than that of bromobenzene. This situation is unprecedented and very improbable. A second possibility is a competitive or

secondary react:lon which is ver-i dependent upon the

concentration of so3 , which produces bromobenzenesulfonic acid more selectively than benzenesulfonic aci.d. ~ere are at least two possible reaction mechanisms which involve secondary reactions. one, which is suggested by various observations reported in the literature 4 •5d

0 involves.the formation of the pyrosulfonic acids in the primary reaction which in turn sulfonates starting aromatic compounds in the secondary reaction at a selectivity favor- ing the production of bromobenzenesulfonic acid.

•.4 126 ·• 'Mechanis.."ll I

AH + 2 ,S---+ ASSH BH +. 2 S--+ BSSH

:'( ,..:l'\,l ASSH +. AH ~ 2 MH ASSH -i· BH ~ ASH + BSH

BSSH + AH ~ ASH + BSH

BSSH + BH -. . 2 BSH

. A second possible mechanism involves the formation of the sulfonic acids in the pri~y reaction, a portion of which rea.cts with more so3 to give the pyrosulfonic acids which ma:y then sulfonate more of the arana~ic starting materfals in reactions analogous to the secondary reactions i.n the first mechanism above.

Mechanism II

AH+ S -.ASH BH + S ----+ BSH

ASH + S --, ASSH

BSH + S _.,...._. BSSH

ASSH + AH--+ 2 ASH ASSH' + BH---+ ApH + BSH BSSH + AH~ ASH + BSH

BSSH + BH ~ 2 BSH

·oata generated artificially .. with certain arbitrary but necessary,asswnptions, indicated that the second 127 mechanism probably more nearly describes the reaction of so3

sulfonations of a~omatic compounds in liquid so2 than mechanism I.

If mechanism II apdli'!",~t•iy d•scribes what occurs in the reaction mtxture, then the behaviour of the plots (Figure 2) of the data in Table 2 may reasonably be interpreted·-as showing a dominati11g primary reaction at high

.< benzene/'bromohenzene ratios.and low S03 concfµltrations in , the regions where these plots show linearity. Deviations from linearity may be then interpreted as the effects of

secondary reactions&,, If this· is 'true, then the relative '. " rate constant ratio may be calculated from the ratios of benzenesulfonic acid and. bromobenzenesulfonic produced and the amounts of benzene and bromobenzene initially ·added to· the reaction mixture and calculating kA/kB by the Ingold equation (1). The isomer distribution resulting fran the ' sulfonation of bromobenzene,under the same reaction conditions as the competitive reactions is necessary for the calculation of the partial rate factors (equations 3, 4, and

6 % E-isomer X (3) = 5 X 20%

% g;:..isomer . Of_ = ~ X (4) 5 40% 128

6 ·o1 m-isomer = X '° X (5) 5 40%

The resul. ts of six expet· :fJnents ate shown in Table 3. It has been suggested 5b.that secondary reactions may have a significant effect on the isomer distribution. It was reported by Cerfontain,_ et ai. 16 that the ortho/para

(16) H. Cerfontain, H.J. Hoffman and A. Telder, Rec. Trav. Chim., 83, 493 (1964). H. cerfontain, :ir~ L. Sixma and t.· Vollbracht, ibid., 82, .,659 (1963). ratio decreased as the reaction progressed in the S\.tlfonatia:i of toluene. This can be rationalized on the basis of the steric hindrance of the phy;ically larger secondary sulfonating species which become more significant.as the . react~on progresses. The data in Table 4 show that in these five experiments, the results.with bromobenzene are consistent with such an interpretation. From the relative rate constant ratio ,,.and the isomer distribution obtained in this investigation, the partia; rate factors were calculated by equations (3, 4, and-5). These values are P:f = O.2O8, Of :::;,O.,OO96and mf = 0.00010. l2S Experimental section 'l.1he apparatus used in' the sulfonation reactions was

•., the same for both the relative rate and isomer dis.trib~tion

·determinations and is shQwn in . J:4.~~1 3.. A t.ypi cal run ' . .~·( "

Fig. 3.--sulfonation apparatus.:

S.n·•,'01ved condensing 100· ml. ,bf so2 (Matheson, anhydrous grade) on a Dry Ice-methanol cold fi.nge:r ir1to the collection vessel., A.,' which is cooled to -20 to -4ooc in a dry· ice cooling bath. '!'he so2 was forced over into the reaction vesselll' B, by a positive nitrogen (Whitmore, 99%, anhydrou~) presstu:e, ai."ldwhile mainta.ining the slight positive pressure of N211 the so3 .in a thin-walled capsule was introduced through the center neck of the reaction flask and the capsule was crushed with a glass stirring rod. Another 400 -ml. of so2 was:collected and introduced into the reaction vessel t-..nd the· so3 was totally dissolved. 'l'he aromatic r~acta.nt (s), bromobenzene for the isomer distribution reactions and mixtures of benzene and bromobenzene £or the competitive rate 130 z:eactions, were introduced into the collection vessel

through t.he side arm, c, and ,so2 collected until· the total volume in the collection vessel was 135 ml. The cooling . .. ' baths were removed from around A and Band both solutions were warmed to reflu.ic tefnpe'l:'ai;:ur~, and stirred as vigorously as possible magnetically. With both solutions stirring and refluxing, nitrogen pressure was applied to force the hydrocarbon solution from the colrection v.,essel into the reaction vessel. This transfer required 5 to a seconds. The temperature was determined by means of the thermocouple, D. The reaction mixture was stirred for an additional _10

minutes and the.n the apparatus was disassembled and the so2 allowed to evaporate. About 200 ml. of ether was added to the product residue and boiled off to remove the last trac~s of so2. The products were then taken ~Pin 200 ml. of ether and extracted four times with 50-ml. portions of water. The collected aqueous_ fractions were then washed twice with ether and titrated with standardized NaOH to pH .7 as. :<, followed by a Leeds an4 Northrup Model 7401 pH meter.

Determination of relative rates.--Aft~r the titration of the products of the competitive sulfonation, the salts were evaporated t.o dryness a~d samples were

1 7 by converted to the sulfonyl chlorid~s heating O. 5 g,."":/ of

(17) J •. s. Parsons, J. Gas. ChromatQ5u_, 5, 254 (1961). 131

these salts with 3 g •. :PC15 in a test tube fitted with a condenser to 110°c for one hour wlth stirring by means of a stirring rod extending through the condenser. Ten ml. of benzene were added through the condense«r and after cooling , .f: , · ·/:" :· ,, 1 the benzene solutj,on was drawn t:,ff afld the excess PCls decomposed by shaking with 5 ml. of water. The benzene

solution was then separated and dried over cac12 and s'amples were injected int.o a Varian model 200 gas chromatograph. The separations wez:'e made on a fo1.1r-foot by 1/4 inch stainless-steel column packed with 4% SE-30 (methyl silicone gum rubber, Varian) on 80-100 mesh Chromosorb w,

and detected,, by thermal conductivity. The :column'_was maintained isothermally at 12s 0 c with a helium flow of 40 ml./min. Integration of the peaks gave the relative amounts of the sulfonyl chlorides. Preparation and separation of the sulfonyl chlorides prepared from known mixtures of_ benzene and'bromobenzenesulfonic acid salts showed that the conversion to the sulfonyl chloride was non-selective under the conditions used.

Isomer distribution determinat!2!!,!.--Bromobenzene. 35 was sulfonated by the above procedure using s o3 • A,liquots of the product salts (after titration) were added to about 40 g. of previously prepared non-~adioactive, isomerically pure p-, o, and m-bromobenzenesulfonic acid salts., A calculated 5% excess of p-toluidine and HCl were added and 132 t-~a.chisomer was, re,:::y,,talli:ted as the p-tolu,idine · salt from 13 to ·21 times as J.c,ng as sufficient rfZ'.ma.ined for: recrystallization. Samples after each recrystallization were taken for subsequ_ent :radi<.\tion These were dried over- -~-ii;;,' 111.ght undel' vacu~--n at f '."'J. 5 tug. wei.ghed into ccintillat.ion viale, dissolve'd in l ml .. H2o, 1 ml.·absolute 18 ______methanol and 15...... ml,. Kinard solution and counted in the (18} F. E. Kinard, ~ev. Sci. In-=9t;_r~,l8, 293 (1957). ------·-- scintillation counter. ,.- Calculation of the _percent. of each isomer formed in

-~1 • the reaction.was made utilizing the specific .activity of the 2::ample, the relative molar quantities of the inactive salts, and the relative size of the.aliquots of the titrated solution used for each isomer.

preparatj._gp of s~:23~.-.-The radioactive sulfur was 35 obta:i.ned.frO£'TI. New England Nuclear as .a2s o4 which was washed into a bealcer containing 2 g~ BaS04 an,d 0.25 ~•

Bacl 2 , anp allowed to digest on a steain plate for at least • ' t.hree days. The radioactive Baso4 was washed s·everal times by decanting the water and evaporated to dryness finally under vacuum at 120°c overnight .. ·.. It was then divided between three_ ..excha.nge reactors which consisted of a filling tube connect:_ing two reservoiz::s one of .-nich had two to three thin-walled capsules sealed to it .. sulfur t.rloxide (5-8 ml.

Sulfan B, Baker and Adamson, Stabilized)•. was introduced into 13:!f

the reactor whic-.h wa~ then placed in a.n ice bath to freeze

the S03. After the so3 was . completely frozen, the reactor was evacuated to ahout 1 Torr, and t..1-iefilling tube sealed. The S03 and Baso .~rar:; ~llowe_d to e:.:chang-e for at least three 4 i ' ,_.,.;; ~ f, days with occasional s tj ./. .. r to filling the capsules,

the so3 was distilled f~~ the e.xcha."'lge reservoir into the seccn~ reservoir which, along with the. thin-walled capsules, were completely rinsed to remove any trace of water or n2so4 • 'l.l1e capsules were individually filled, :erozen, seal'ed off

and removed from .the rea.ctor •. · The precise a.mount of so3 in . ' . each capsule ·could not. be c:ontrolJ.ed under these conditions but was deteJ;mined by the amount of acid pi:·oducts titrated ,with :the·' standard NaOH.

Purification of henzen,e and br(!!!_obenzerie. --The .ben•- zene (reagent grade) was distilled from sodium metal throui:;h ,, a 4..;ft. helices packed, vacuum- ja;cketed colunm and a c-enter , " . 0 cut collected· at 72-72.s c (unc~rrected) and stored in the·

col.d in a tightly stoppered J'?o~tle. Bromobe.nzene (reagent 9~rade) was distilled through the same column and sj.milarly stored. ,. Samples of the benzene a??,.dbromobenzene were inJect..:ed into a Varian model 200 gas chromatograph. Contamination was not .apparent at the mpXimum ~ensitivity of t."le instrument •

.~ 134 Preparation of inactive p-, 2:.-• and m-bromobenzene-

,!_Ulfonate isomers. Large qu~ntit~es of each isaner were prepared by the method of Mee~ei\n. et al. , 19 which involved

(19) H. Mee~in, G. Dif-~~,, ,1t,.~~-'~Ji1ner,K. ·Hafner, F. Mensch and o. St~infurt·; Ber., .2Q, 841- (1957). the diazotization of the respective bromoaniline, d_~composition of. the diazonilJitt chloride in so2 saturated acetic acid, and hydrolysis of the su.lfonyl ~loridE!. A typical run involved ~issolving 2oo·g. (1.16 moles) of _e-bromoaniline- (Aldrich Chemical Co .. ) in 375 ml. of ethyl ether and, while maintaining· '.the ~emperature at less than 10°c·, addj.ng 410 ml. of concentrated hydrochloric acid. To the p-bromoaniline hydroea11loride was then added, with cooling and vigorous stirring, 85 g.' (1.3 mo~es) of NaN02 in 175 ml. of water. During· this addition; 900 ml. of glacial acetic acid, cont,aining 37 g. of Cu2Cl2•2H20 (Baker•s c. P. grade) dissolved in a minimum amount o~ water, was satur~ted with sulfur dioxide (war surplus, grade unknown) in an 8-l. 'jar. The diazonium chloride was poured into the chilled aeetic acid-sulfur dioxide solution as rapidly as pos&ible with vigorous stirring and/or use of a compressed air stream to control the profuse foaming. Bubbling of so2 through the reaction mixture was continued for an hour. _ Theo- and p-bromobenzenesulfonyl chlorides solidi- -- - ,·\ . fied upon addition of 1500 g. of ice to the sulfonation products. ·T?ese were c~llectedby filtration,,washed.with 135 cold water and hydrolyzed with potassium hydroxide (Allied Chemical., re.agent). The ~~bromobenzenesulfonyl chloride was a liquid and was separated., after the addition of the ice, from the . , reaction mixture by extraction with five 300-ml. portions of ethyl ether. The ether was evaporated and the sulfonyl chloride was hydrolyzed with potassium hydroxide. The potassium salt: of the meta isomer was not conveniently recrystallizable so it was converted to the barium salt by the addition of barium chloride. The solutions of each-of the bromobenzenesulfonic

and acid salts were treated with .·• Norite and filtered crystallized readily upon cooli:ng. The salts were collected by vacuum filtration and recrystallized from 25% ethanol in ·, •;,i water until pure. The overall yields for the three isomers were 49% of theoretical for the meta isomer, 59% for the para and 51% for the ortho isomers. The S-benzylisothiouronium derivatives 20 _of the

(20) E. Chambers and G. w. watt, !l_. Org. Ch~., !, 376 (1941). isomeric bromobenzenesulfonic acid salts were ·-prepared by disso~ving '·o. 25 g. of the salt in ·water and adding 1 ml. of an acr~eous 10% s-benzylisothiouroniurn chlor~de solution.

The pro~uct was recrystallized from water three or four times •. The melting points (corrected) of the derivatives 136. 22 were p-, m.p. 175.5-177 ·(lit. 110, 21 177~a ), 2-, m.p.

(21) N. D. Cheronis and J.B. Entrikin, "Semimicro QUalitati ve organic Ar1alys1~·, •• 2d ed., Interscience Publishers, Inc., New York, 1-1. Y., 1957, p. 696. (22) J. c. Robert.son, JDdctt,ral .Ois~ertation, Brigham Young University, 1962.

147-8 (lit. 148.5-151 22 ), m- m.p. 130.5-131.5 (lit. 132-

133 22. ) • 13'7.

Table 1.--variations of the Relative Rates with Initia.l Benzene/Branobenz~ne Ratios

Initial Ra.tic k¢H/k¢Br /dH/¢Br

2.1888 22.37 i.9875 72.31 1.9554 17.09 1.9553 11.90

1.0307 20.06 1.0002 34.97 0.9899 17.12 0.9847 15.76 0.9841 39.0l 0.9812 24.53 0.9707 42. t3 0.5072 47.60 0.4999 36.54 · 0.4972 12.93 . 0.4939 16.60 o.4837 18.07

0.1378 30.91 0.1266 29.29 0.1252 57.90 o.11a2 21.47 138

Table 2.--Moles of the :f!"'inal Products in· the Competitive Sulfonation of Benzene and Bromobenzene - .. G. ~~ Separation Run Initiai S03 No. Ratio ¢so3H·. srtsst>3k ¢S03H/Br¢s03 Mmoles Mmoles Mftloles . Ratio

44 0.9812 15.05 0.70 21.50 15.75 45 0.9841 22.01 . 0.66 33.30 22.64 46 1.0002 7.54 0.23 32.78 7.77 41 1.0134 29.52 0.03 984000 29.55 48 0.9899 35.50 2.67 13.24 38.17 68 0.9707 8.90 0.22 40.45 9.12 71 1.0307 13.00 0.68 19.12 13.68 Avg1 0.9957 •·

49 1.9553 34.25 1.73 19.80 35.98 50 1. 95S,4 28.64 o.98 29.22 29.26 51 .. 2.°1888 16.87 0.36 46.86 17.23 69 1.9875 1.00 o.os 140.00 ·. 7.05 Avg1 2.0218

52 0.4972 30.42 6.59 4.62 37.01 53 0.4999 11.40 o. 70. 16.29 12.10 54 0.4939 35.25 6.57- 5.37 41.82 55 0.5072 2~17. 0.09 24.11 2.26 65 0.4837, 30.13 5.15 s.as 35.28 Avg1 0.4964 ·.~ .. 60 0.1266 19.45 8.57 2.27 28.02 61 0.1378 11.56 3~45 3.35 15.01 62 0.1182 22.33 16c85 1.33 39.18 66 0.1~52 s.oa 0.97 5.24 6.05 Avga 0.127Q ·139

Table 3.--Isomer Distribution for the Reaction of Sulfur Trioxide apq Bromobenzene

--;t.Pa1·centagGs Para ortho ~-- Meta

98.72 1.04 ·0.24 98.87 0.97 o.J.6 98.94 0.90 0.16- 99.04 o.a,2 0.15 98.88 0.96 0.16 99.11 0.76 0.13 98.93 t 0.14 0.91 ± 0.04• 0.1s ± 0.02

Table 4.--secondary Effects on the Isomer Distribution

Ml. ¢Br· Or tho/Para Ratio

23.26 10 0.0138 23.84 10 0.0139 22.32 10 0.0139 8.16 20 0.0145 34.22 5 0.0125 140 50------, \+ . Initial Ratios . -:.:: ¢H/¢Br = 2.0·· 1.0 ··:·· . ~:~:;: o.s \ · ·Q ¢H/¢B~ = 0.1 - 40 0

0 0

30 f· -:-

Id .::,• -~ it, \ A 0•

t,• 0 20 -,-I -:i:: Pl 0 0 ·~ k A

S·Pl 0 .0

~ :t ,. 10

-"□ ~□ ------□--, -- □- 0 10 20 30 40 i' S03 mmoles

Fig • . 1 •. ~-variations in the ratio 6f the product ~ulfonic acids with s~lfur trioxi'de concentrations as (letepnined by gas chromatography. Initial Benzene/Bromobenzene Ratios / ,_,,- 6 3o~ -1-¢H/fjBr = · 2. o p _6 ..- 0 ¢H/¢Br = 1.0 ~••- 6 ¢H/¢Br = 0.5 ·25~ . O ¢H/¢Br = 0.1 . ✓ ------□-

ti) 20 ./·. 0 D

,-j0,) , I / / 'j . V . D .. :r: 15 /'6.. t . o··/ .. ~- / . ~ . /f). .,..o 10 0,0/ /0 y A-1/ .. ~- .A ·A Sl /40' ~ .A / ~- -□ . ~- ' . 0 6 . A O O . . ::=--.--. . ----· •-• --· -,· I/ (',,-e:;;::::- D -:Q-0-;--'->- :~ --:-- 0 5 10 15 20 2.5 30 35 40 45 50 so Mmoles 3 .... Fig •. 2.--Moles of sulfonic acid products/sulfur trioxide (G. c. data)...... s:,. THE REACTIVITY AND SELECTIVI1'Y OF THE REACTION OF SULFUR TRIOXIDE AND BROMOBENZENE

Sullivan E. Blau Departrtilht bf dtii:nrtistry Ph.D. Degree, May 1970

ABSTRACT

The sulfonation of bromobenzene with sulfur trioxide in liquid sulfur dioxide at -12 to -13° c. was studied to determine its relative reactivity. The relative rate constant ratio, kbenzenefkbromobenzene, was found to be 28.6 in a competitive reaction between benzene and bromo benzene. The isomer distribution was found to be 0.15 ± 0.02% for the meta, 0.91 ± 0.04%'for the ortho and 98.93 :I: 0.14% for the para isomers. The partial rate factors were calculated to be Pf= 0.208, Of= 0.0096 and mf = 0.00018. The relative rate data e>,:hibited a distinctive behaviour previously noted in related systems, possibly due to secondary reactions. Data, generated artiticially, which showed similar behaviour, indicate, that possibly the primary and secondary reactions both produce the final products in competition with each other at significantly different selectivities.