Eastern Illinois University The Keep
Masters Theses Student Theses & Publications
1981 Analysis of Coal Fractions Using Dibromotriphenyl-Phosphorane: A Selective Ether Cleaving Reagent Michael L. Ballard Eastern Illinois University This research is a product of the graduate program in Chemistry at Eastern Illinois University. Find out more about the program.
Recommended Citation Ballard, Michael L., "Analysis of Coal Fractions Using Dibromotriphenyl-Phosphorane: A Selective Ether Cleaving Reagent" (1981). Masters Theses. 2969. https://thekeep.eiu.edu/theses/2969
This is brought to you for free and open access by the Student Theses & Publications at The Keep. It has been accepted for inclusion in Masters Theses by an authorized administrator of The Keep. For more information, please contact [email protected]. THESIS REPRODUCTION CERTIFICATE
TO: Graduate Degree Candidates who have written formal theses.
SUBJECT: Permission to reproduce theses.
The University Library is receiving a number of requests from other institutions asking permission to reproduce dissertations for inclusion in their library holdings. Although no copyright laws are involved, we feel that professional courtesy demands that permission be obtained from the author before we allow theses to be copied.
Please sign one of the following statements:
Booth Library of Eastern Illinois University has my permission to lend my thesis to a reputable college or university for the purpose of copying it for inclusion in that institution's library or research holdings.
Date
I respectfully request Booth Library of Eastern Illinois University not
allow my thesis be reproduced because -��
Date Author
m Analysis of Coal Fractions Using Dibromotriphenyl-
Phospborane: A Selective Ether Cleaving Reagent (TITLE)
BY
Michael L. Ballard
THESIS
SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF
Master of Science IN THE GRADUATE SCHOOL, EASTERN ILLINOIS UNIVERSITY
CHARLESTON, ILLINOIS
1981 YEAR
I HEREBY RECOMMEND THIS THESIS BE ACCEPTED AS FULFILLING THIS PART OF THE GRADUATE DEGREE CITED ABOVE
DATE
J/__.a1DATE /ti ANALYSIS OF COAL FRACTIONS USING DIBROMOTRIPHENYLPHOSPHORANE:
A SELECTIVE CLEAVING REAGENT ETHER
by
Michael L. Ballard i
Title of Thesis: Analysis of Coal Fractions Us ing Dibromotriphenyl Phosphorane: A Selective Ether Cleaving Reagent
Michael L. Ball ard, Master of Science, Au gust 1981
Thesis directed by: Dr. David H. Buchanan
Abs tract
0 Dibromotriphenylphosphorane (DTP), in acetonitrile at 4o c, was found to cleave benzyl ph�nyl ether in preference to di-alkyl or ot her aryl alkyl ethers . Competition between 10 mmoles benzyl phenyl ether and 10 mmoles di-n-octyl ether for 0.92 mmo le of DTP
1.3% gave 48% cl eavage of benzyl phenyl ether and cleavage of di-n-octyl ether based on DTP . Reaction of benzyl phenyl ether and ° di-n-octyl ether separately with DTP at 40 c produced 80% and 8% cleavage, respectively. Reactions of DTP with diphenyl ether, butyl
00 C phenyl ether, phenyl cyclohexyl ether and benzyl benzoate at 4 produced less than 3% cleavage in all cases.
Reaction of the toluene-insolubl e, pyridine-soluble fract ion of Il linois No . coal = 1040) with DTP produced fractions 5 (MWn containing phosphorus and bromine with lower number average molecular
Treatment of the DTP-reacted fractions with 10% HC1/CH CN weights. 3 produced one fract ion of which the of the pyridine-soluble portion MWn was 3150-, and another fraction the of which was 2239, \\Tith 67.3% MWn of the original carbon re�overed (from the total DTP -reacted fract ion) and 63.6% of the original carbon recovered (b ased on the original coal fraction) , respectively. Th e solubility of the DTP-reacted fractions in pyridine decreas ed.
From IR spectra, elemental and molecular weight anal ys e s and ii
the reactions of model compounds, it is proposed that DTP is cleaving benzyl ether linkages in the pyridine-soluble, toluene insoluble fraction of Illinois No._ 5-;coal to form reactive fragments which upon further manipulation react to form higher molecular weight aggregates. iii
Acknowledgement
I would like to thank my Research Advisor, Dr. David H.
Buchanan, for his help and guidance, the Faculty and Staff of the.Chemistry Department at Eastern Illinois University, and most of all my wife Peg, whose support, understanding and love made the whole project possible. iv
TABLE OF
CONTENTS
bstract •••••••••• •••••••• ••••• • •• • • •••••• ••• A
Aclmowledgments ••••• ••••• •••••••• • ••••• •••• • • iii
List of Figures •••••• ••• •••••••••• ••• •••••• •• v
of T bles ••••••• ••••• ••••••• • •••••••• ••• List a vi
I. Introduction •••• ••••••• •••••••• •••••• • ••• •••• 1
Experimental •• •••••••••••• •••• •••••••••• ••• •• II. 12 Reactions of Model Compounds ••••• ••••• • • 12
Extraction of Whole Coal •• • •• ••• •• •••••• 30
Reactions of Asphaltols •••••••••• •••••• 35
III. Results and Discussion • •• ••• ••••••• ••••• ••••• 39
Results of Model Compounds •••• •••••• •••• 39
Results of Asphaltols ••• ••• ••••••••••••• 59
IV. Conclusions • •• • •• •••••• • •• •••••••••• ••••• ••• • 90
References • •••• ••• ••• •••• • ••• • •• •••••••• ••••• 92
Vita ••••• • • •• ••• ••• ••• •••• • •••••••• •••••••••• 94 v
List of Figures Figure No. Page
1 Hypothetical Coal Model ••••••••••••••••••••••• 6 2 Results of Cleavage Rxns. on Asphaltols (Mayo}. 9
3 Graph of w./w vs. A.IA - Samples I, II and •• 46 1 s V, Rxn. I.t(2'
4 Graph of w./w vs. A.IA - Samples III IV, ••• 7 1 s 1 s & 4 Rxn. I.F(2)
5a&b IR Spectrum of Asphaltol 252-B •••••••••••••••• 61,62
6a&b IR Spectrum of III.A(1)-a ••••••••••••••••••••• 66,67
7a&b IR Spectrum of III.A(4)-a ••••••••••••••••••••• 68,69
•••••••••••••••••••• 71 8 NMR Spectrum of III.A(1 )-a
9 a & b IR spectrum of III.A(1)-h ••••••••••••••••••••• 74,75
••••••••••••••••••••• 76,77 10 a & b IR Spectrum of III.A(4)-h
11 a& b IR Spectrum of III.A(2)-a ••••••••••••••••••••• 79,80
12 a& b IR Spectrum of III.A(2)-b ••••••••••••••••••••• 82,83
III.A(2)-b •••••••••••••••••••• 84 13 NMR Spectrum of vi
List of Tables
Table No. 13! 1 Ultimate Analyses of 4 Di.£ferent Rank Coals •••• •• 2 ° 2 Ether Cleavag e Rxns. At 81 c in CH CN • •••• ••••• •• 42 3
Experimental Results of Standardization of • •• 3 Rxn 45 I.F{2}
Quantitative Analysis of . I.F{2} •• ••••••••••• 48 4 Rxn
5 Quantitative Analysis of I.A{B} ••• •••••••••• Rxn. 50
Quantitative Analysis of . I.D(5} •••••• •••••• • 51 6 Rxn
Experimental Results of Standardization of • •• 51 7 Rxn I.G{2}
Quantitative Analysis of . I.G{2} ••••••••••••• · 8 Rxn 51
Experiment l Results of standardization of . • • 52 9 a Rxn I.H{1}
10 Quantitative Analysis of I.H{1} ••••••••• • • •• 52 Rxn.
11 � Experimental Rasul ts of Standardization of . • • 5 Rxn 3 I.J{1}
2 Quantitative Analysis of I.J(1} • ••• •••• ••••• 54 1 Rxn.
1.3· Experimental Results of Standardization of •• Rxn. 54 I.K{1}
4 Quantitative Analysis of I. {1 •• •••••• ••••• 1 Rxn. K ) 55
15 Rxns. of DTP With Model Compounds for 2 Hours •••• 58 at 40°c in CH CN 3
Ultimate Analyses of Whole and Extracted Coal •••• 59 16 Fractions
Absorption of • 17 Prominent IR Bands Asphaltol 252-B 60 vii
( List of Tables cont. ) Table No.
18 Ultimate .Analyses of Asphaltols After Rxn• ••••• • with In'P
19 Phosphorus Corrected .Analyses of In'P-Reacted . '• 64 . .. Asphaltols
20 Phosphorus Corrected .Analyses of DTP-Reacted ••• • 73
and Hydrolyzed Asphaltols 21 252-B III.A(2)-a Ultimate Analyses of Asphaltol and 81 1
Introduction
Coal is the end product of millions of years of geologic and chemical alterations on plant and tree matter deposited in swamps and marshes under conditions allowing only partial decomposition of the material. These geologic and chemical processes
' made coal into such a complex mixture of organic, inorganic and volatile components that scientists have been unable to completely decipher its structure at this time. As the search for new exploitable energy resources intensifies, more and more attention is being focused on coal. Almost one third of the world's known coal reserves are U.S. located in North America, with 97% of these lying in the
Thus, the solution to the energy problem may lie in the unlocking of the mystery that surrounds the structure of coal. 4 The ultimate analyses of types or ranks of coal {dry ash fr ee basis ) are shown in Table 1 . An ultimate analysis expresses the composition of coal in percentages of carbon, hydrogen, nitrogen, sulfur and oxygen. The results in Table 1 have been corrected for ash content. The terms high-volatile and low- volatile refer to the amount of gases and low-boiling components trapped within the structure of coal. These volatile components are determined by measuring the weight loss of a coal sample after heating to in a platinum crucible for minutes From the 950° 7 ! figures in Table 1 it is observed that on a weight percent basis oxygen is the second most abundant element in coal after carbon Empirical Formula c H 0 N s H/C
Lignite 65.20 4.66 27.43 1.77 0.98 o.85 C H 0 I'\) 178 152 56N4S
High - volatile Bituminous 80.50 6.40 7.2 1.o6 3.58 0.94 4 C157H150°11N2S Low-volatile Bituminous 82.69 .29 9.89 .8 23 0.77 H 0 .5 0 9 1. C179 138 16N2S Anthracite 92.76 .4 2.14 o.88 0.73 o.45 1H 0 3 9 C29 153 6N3S
Table 1
Ultimate Analyses f Different (daf) o 4 Rank Coals Weight Per Cent oxygen is the third most abundant element on atom percent ( an basis . A considerable amount of work has been done in an effort ) to deternJ.ne the role that oxygen plays the structure of coal. in Statistical, chemical and spectroscopic methods have accounted for approximately of the oxygen in bituminous coal in.the form of 70% hydroxyl and carbonyl groups� · This leaves approximately 30% of the oxygen present in coal with unassigned functionality. More
more evidence is pointing to the conclusion that the remaining and oxygen is present in the form of ether linkages. Infrared spectra do not indicate unmistakably the presence of ether oxygen, but its presence may be inferred by the variation of the spectra with the rank of the coal. Also, ethers tend to be inert in most chemical reactions and therefore could not be detected like phenols or carboxylic acids.
work of Ignasiak et has shown that reactions of coals The �-6 ali.d-·coal fractions with Na in liquid NH and 2) K/THF/naphthalene 1) 3 followed by alkylation produced significant decreases in the average molecular weights of the coal samples. Both 1) and 2) are known to cleave di-aryl and benzyl ethers. A mechanism is shown
F.quations in (1 )-(4). other workers have reported similar results.
J. + ArOAr Na ArOAr + Na (1 ) liq .& • NH3 - + ArO (2) ArOAr Ar • + Na + Na (3) Ar Ar-
- + - + + ArOH (4) Ar Aro ..solvent ArH 4
7 Wachowska reported that treatment of several coals with
K/THF/naphthalene lowered the molecular weight of the coal and increased its solubility in benzene. This is evidence of the degradation of the coal structure by cleavage of ether linkages between various structural fragments in gen on coal. Hydro ati of coals room temperature induced the splitting ether at of linkages accompanied by dehydration and increased benzene solubility� 9 Selective oxidation results support the theory that the organic portion of coal consists of aromatic units connected by aliphatic and ether link ges was found that the size of a . It these aromatic units increased directly with the rank of the co , lignite composed chiefly of benzene rings and anthracite al with containing phenanthrene as the major structural unit. Evidence 0 for ether oxygen in coal was also presented by Yoshii and Satou! The authors isolated a crystalline compound from a pyridine extract of Sorachi coal which contained an alicyclic ring or methyl-substituted pentacyclic ring, an ether th 2-3 condensed wi aromatic rings and a mixture of substituted anthracenes. This showed the presence of methylene bridges and oxygen as an ether 1 atom. Mayo and Kirshen � · after comparison of fractionated pyridine extracts and solvent refined coal (from Illinois #6 coal) using gel-permeation chromatography, concluded that the solvent refin�ng process was ccessfu for the most part su l in cleaving ether linkages and the the organ removing bulk of ic oxygen and found in coal without affecting the concentration eUlfur
of phenolic-type groups.
um t v Taken acc ula i ely these data strongly suggest that the probable structure of typical medium rank bituminous coal a (such as Illinois #6 coal) consists of small clusters of condensed
aromatic (anthr cenes , phenanthrenes, etc.), aromatic rings a heterocycles (benzofuran, indole and quinoline e.g. ) and hydro
aromatic structures (such as tetralin linked together by short ) ether, methylene and methtne bridge These small clusters are s. ·"in turn ·linked and cross-linked in much the same manner to forrn
- dimension array of larger clusters with an average molecular a 3 al weight in the region of 3000-5000. One of the better known models 2 suggested for bituminous coal is that proposed by Wiser! This model (shown in Figure 1) illustrat s the various types of structures e that are thought to be present in coal. Note the various types
of and use of short aliph tic in ether'linkages the a chains 13 holding the structure together. Chakrabartty and Kretschrner
claim the presence of aliphatic groups (n-propyl and n-butyl ) by their work with sodium hypohalite oxidations of coals.
The model in Figure 1 also takes into account the presence of
heteroatoms such as nitrogen and in coal. sulfur. With thi8'�elatively simple but well-founded concept cf the
structure of coal we can now begin to look at further experiments
which would tell us more about the structure of coal, the reactivity
and importance of the ether linkages in that structure and how 6
. ------·------· H I H 0 S� . I I I H 0 H H /H I ,,- � �I I H" H 0 H, .... H \ .... I ' ' o- ...... I s I ,,.,:;-0-o... c' H H ..... � 'H w-'- # ..- 2 - o 'H I I b,- H c H H-C-H H....- 'H ,-c,;-m,H..... �o o 9H 2 ' I H - - H H H c H �2 � I H-C-Hl 1 H """",..� - -H 2 ..... cH 9 H H H i H , � , � �-oH,O- ..... H I I 1 N H I, 0 H H '- ' H CCO'# 1-t-twH H Hf "HW'- 6t I f _.. 2 I I HH2 I H I I ,....o H-C-H s 'ef-...... l'H H 2 H 2 I H ,.. H-C-H I 0 'a-=DN H w" I _,p,H H-C- H, I C ...- H H-C-HI I I I H �z H H-C-H H I � H H .... I " I H I I _....o H H-C-H I 'H 'o , o H H H H a-:oI H H H o' I I \ / ...... , \ I \I � ...... 2 H, _...... c, � H � o....,...,Q- ,...H2 Q- W" 0 C 'H I \ I \ ..-H W"' 'H H H H H ...-H2 I H
Figure 1 Hypothetical model illustrating the structural groups and
connecting bridges thought to be present bituminous coal.1 2 in
these might be broken or cleaved to depolymerize coal. This
depolymerization would lower the average molecular weight of
coal and increase its solubility �n e�traction solvents. Any
chemical experi.ment performed on coal is innnediately complicated
by the fact that coal has solubility most if not all limited in ( ) solvents. Furthermore, even when finely ground the pore structure 1
of coal is such that diffusion into the inner recesses of the 7
coal particle is very time dependent. A few workers avoided this
pitfall by working with solvent extracted fractions of coal
which are representative of the whole coal. The asphaltol or
( ) TIPS toluene-insoluble pyridine-soluble fraction of· coal is one example and represents about of the original coal. 1Cf%, The classical method of cleaving ethers is treatment of the
ether with strong, hot, concentrated hydrohalic acid such as a HBr (see :Equations (5')-(7)). This mechanism involves or HI14-15 HBr R-0-R R- -R + Br (5) '> � H I - � - R-�R B R-OH + (6) <- )' RBr HBr (7 ) R-OH RBr + � H2o
protonation of the ether oxygen then SN2 attack by the halide ion to yield one mole of alcohol and one mole of alkyl halide.
The alcohol produced reacts further with the acid to produce
a second mole of alkyl halide and a mole of water.
Another method of cleaving ethers which was mentioned earlier
is the use of Na or K in liquid solution. This method is NH3 effective for cleavage of aryl and benzyl ethers but is relatively
inert to alkyl ethers.
A third method of cleaving ethers 16 developed recently uses
trimethylsilyliodide. This reagent was successful in quantitatively cleaving a wide varie�y of di-alkyl, alkyl-aryl and t-butyl ethers B
at room temperature. Equation (8) illustrates the mechanism of the reaction. '
' Me SiI (8) R-0-R 3 � R-Q-'��-SiMe + I 3 CHC13 l ' R-OH + R-0-SiMe R- ·1 J
In a study on the effect of cleavage reactions on the solubility and molecular weight of asphaltols from Illinois #6 coal, Mayo et !!,�7 investigated the above reactions and several others. Figure 2 lists some of the results from this study. The numbers C for weight percent and atom percent C are based on 100 C atoms of starting material and the numbers for the products indicate
C recovered on a weight percent and atom percent basis. Reaction
6 of asphaltol 1-D with Me3SiI in CHC13 yielded product -D with a molecular weight 39.2% lower than the starting asphaltol.
Reaction of 1-D (different sample) with pyridine hydroiodide . 63% (Py HI) in pyridine produced 15-A which was lower in molecular weight than the original asphaltol. Pyridine hydroiodide was used because it was thought that it would be an acid in pyridine in which the asphaltol was soluble� Asphaltol 17-B was treated with HBr in pyridine to give 24-B and a 37% decrease in molecular weight. Asphaltol 28-A (which was very similar to
17-B) with a was treated HI in pyridine yielding 6-A and 60% 35-A reduction in molecular weight. Asphaltol when treated with 9
6-D 90/.738/.109 113/760/ -
Asph 1 00/.744/.113 1-D 100/1250/3.2 Py•H 6d room temp in Py
15-A 90/.746/.133 102/462/0.6
,HBr Asph 100/.787/.125 24-B 81/.790/.133 17-B 100/980/ - sIJ 83.4/ 616/ - 68h
Asph 100/.782/.140 HI 6-A 84/.787/ - 28-A 100/1090/ - 89/430/ - Py room temp 68h
Na/NH;/BuNH Asph 100/.781/.118 2 24-A 91/.897/.128 ° ) 35-A 100/1380/1.9 -35 93/790/4.5 Sh
% _s;1I;c Key: Sample Atom O/C Number % / Wt C/Mn/ eq. OH/mole
Figure 2
Results of cleavage reactions on asphaltols from Illinois #6 coal. 10
Na/NH /BuNH od 3 2 · pr uced 24-A which was·42,8�·lower in molecular weight than five the original asphaltol. In all of these examples,
80-90% of the original carbon was recovered. These results show that it is possible to degrade the asphaltol structure under mild conditions to i ni an produce s g fic t decreases (37-63%) in molecular weight and good recovery of the original carbon. They also show that reactions that are known to cleave ethers react with coal extracts to break down the structures of these extracts, in effect depolymerizing them to a certain extent.
This thesis describes work intended to find further reagents ; for ether cleavage in coals. After a thorough literature investigation a relatively new reagent, dibromotriphenylphosphorane
(IYI'P), was chosen. It had been reported that the compound was 18 19 useful for conversion of alcohols to alkyl halides - and 20 Burton and Koppes had induced cleavage of several esters with 21 IYI'P. In 1972 Anderson and Freenor showed that DTP cleaved primary and nd alkyl-t-bntyl seco ary symmetrical ethers, ethers and d 60-99% a cyclic ether with yiel s of ( di-isopropyl, di-allyl and isobutyl t-butyl ethers were cleaved the least with 67%,
and 65% cleavage 69% Jrespectively ) . Most of these reactions ° benzonitrile. were carried out at temperatures from 110-120 in
It was reported that acetonitrile was unsuitable as a reaction solvent. These reaction conditions were too vigorous for our ° use since it was desired that temperatures be kept below 110 to insure that no thermal rearrangements were observed. The 11
preparation of is shown in F.qud.ion (') and the mechanism Ul'P ether cleavage using is shown in F.quations of Ul'P (10)-(12). PhCN Br + Ph (9) Ph3P Br 2 » Ph-�''' ,�Ph Br
PhCN of'. .. •. + ROR - + Ph3PBr2 � R-o:!R' Br- {10) I Ph3PBr + - (11 ·�+ - + RBr + Br ) R-�R Br- >' Ph3P-O-R I Ph3PBr
{12) + Br - + ·��- RBr Ph3P-()!.R � Ph3P..o
It is the purpose of this work to:
(a) Investigate the ether cleaving capabilities of in Ul'P acetonitrile at 81°, (b) Determine the usefulness of the reagent under mild conditions
(<5o0), (c) Determine the selectivity (if of for particular any) Ul'P ethers, and
(d) After suitable investigation the reagent with model of ethers (i.e. di-alkyl, alkyl-aryl, di-aryl and benzyl) to apply the results obtained to reactions ·of with asphaltol of Ul'P an Illinois coal. #5 12
Eiq?erimental Section
General: Melting points were taken on a Thomas-Hoover Uni-melt apparatus and are reported uncorrected. Distillations were performed fractionally using a 150 x 10 mm vacuum-jacketed Vigreaux column unless otherwise noted. analyses ere made with a VPC w Aerograph Model 920 using glass columns (0 .25 in m); Varian two x 2 Silicone GP88 (S:&-30) on Chromasorb Wand 2) UCON 50LB 550 on 1) Chromasorb w. Identities of individual peaks were assigned by comparison of retention times and by co-injection with authentic samples on both columns. A Varian T- 0 spectrometer was 6 NMR to obtain 1 spectra with tetramethylsilane as internal used H NMR an standard. A Perkin-Elmer Model 337 Infrared spectrometer was used to obtain infrared spectra. The infrared spectra of the coal and
fractions were taken as KBr pellets using of KBr coal 2 mg and 2 mg of coal fraction for each spectrum.
Triphenylphosphine was purchased from F.a.stman and was used without further purification. Acetonitrile, also obtained from
Eastman, was dried and distilled from CaH. Bromine was purchased from Fisher Scientific Products, Inc. and was used as obtained. Ethers used in these procedures were the i e members of the Author's syn s z d by research group or purchased from s�andard sources.
I.B(1) �thesis of dibromotriphenylph�suhorane (DTP): The 19 method of Schaefer and Higgins was followed using a 250 mL round-bottom flask, N atmosphere, sware, g 2 oven-dried glas 26.4 16.2 (0.114 mole) Br (0.114 mole) triphenylphosphine, g 2 and 13
150 mL dry CH3CN. Bromine was added dropt.ise to the stirred, cooled mixture of triphenylphosphine in over a 15 min CH3CN ute period, maintaining internal temperature at 0-5°c. White crystals formed when the addition was approximately 50% completed . The solid was isolated by filtration and yielded 41.o g 85 yield ( % ) of off white, slightly wet c stals 147-50°). o literature - ry (MP N value for the melting point has been reported. Attempts to further purify the material were unsuccessful and resulted in ° formation of a white powder (MP 153-4 ). ·A sample of the white powder was sent to Galbraith Labs and the following analysis was obtained: %C 77 .7, %H 5.59, %P 11.0 theoretical values ( for 51.2, 3.58, 7.29). An infrared spectrum DTP: %C %H %P of the white powder (#8416) showed strong absorptions at 11 30 1 and 1160 cm- unlike that of triphenylphosphine but characteristic of the stretching frequency of triphenylphosphine oxide P=O (MP 1 4 5°) which has a theoretical elemental analysis of %C 77 .7, 5 -
5.39 and • It was concluded that the white powder %H %P 11.1 was triphenylphosphine oxide . After this attempt at isolation and purification of the it was decided to prepare fresh DTP material for each reaction and use it in situ. ---
I .D( 2) Re '�tion of DTP with di-U:.odyl ether: The method of 21 d using a 3-neck 100 mL flask, Anderson and Freenor was followe oven-dried glassware, N atmosphere, CH CN, 3.1476 g mmoles 2 40 mL 3 (12 ) triphenylphosphine, 1 .9 12 mmoles Br and 2.3229 (10 g ( ) 2 g mmoles) 14
di-n-octyl ether. Bromine was added dr op�se to the mixture of
triphenylphosphine in CH CN which was stirred and cooled in 3 an ice/H 0 bath. After the addition was completed, the mixture was 2 allowed to warm to ambient temperature and the di-n-octyl ether
was added in one portion. The mixture was heated by oil bath
, 0 The to reflux (81 ) and kept at that temperature for 16 hours. reaction flask was then equipped for simple distillation and the
bulk of the solvent removed at atmospheric pressure (Bp 79-81°).
The remainder of the material was then distilled fractionally
with two fractions collected at 43-45°/0.5 mm (1 .66 9 g) and 9 ° 100-101 /0.45 mm (0.5720 g) . Analysis of fraction #1 by VPC
(SE-30, 100°) showed a small solvent peak and a large peak with
a retention time similar to 1-bromooctane. Co-injection of
reagent 1-bromooctane (Eastman) with a sample of fraction #1
confirmed that the bulk of fraction #1 (1.5 g, 7.8 mmoles)
was 1-bromooctane. Injection of fraction under the same #2 conditions produced a sharp spike identified as 1-bromooctane. ° Another chromatogram of fraction #2 (SE-30, 150 ) produced the
1-bromooctane peak and a much larger peak representing a high- boiling component. Comparison of retention times of the second
peak with di-n-octyl ether along with co-injection of the two materials showed that the high-boiling component was di-n-octyl ether. Fraction #2 represented approximately 1.8 mmoles di-n-octyl ether recovered from the reaction and o.8 mmoles of 1-bromooctane produced . Altogether mmoles of 1-bromooctane (43% yield) 8.6 15
were obtained in the cleavage reaction . 'Approximately 18% of the ether was recovered.
The dark solid residue that remained in the di stillation flask was taken up in hot CH CN, treated with Darco, filtered 3 through Celite and cooled. A first crop of white needles ° t (1. 8520 g, MP 155-7 ) was ob ained. A second and third crop ° ° of crystals produced 0.3928 g 155-8 ) and 0.3740 g 155-7 ) (MP (MP J respectively. These were identified by infrared and NMR 22 spectroscopy (comparison with literature spectra ) and melting point as triphenylphosphine oxide. The total triphenylphosphine oxide obtained from the reaction residue was 2.6188 g (9.4 mmoles) for a 78.5% yield. This figure is not necessarily an accurate
indication of the a mount o J f • reacten' di-n-oct"'�l ether however , since reacts with H 0 to form triphenylphosphine oxide (this DTP 2 method was used to prepare actual samples of triphenylphosphine 23 oxide for reference ) .
Reaction of with ether: e a I.D(J) DTP di-n-octzl DTP was pr p red as before using 5.2506 g ( 20 mmoles ) triphenylphosphine, 3.2 g
( 20 mmoles ) Br and 60 mL CH CN. After addition of the Br the 2 3 2 ° mixture was allowed to warm to ambient temperature (25 ) and
2.4347 g (10�04 mmoles) di-n-octyl ether were added in one portion.
The mixture was heated to reflux and maintained there for 17 1.7 hours. Distillation work-up gave g 1-bromooctane (8.8 mmoles, (0.8 8%):. 44% yield) and 0.18 g di-n-octyl ether n1�oles, 16
Reaction of DTP I.D(4) DTP with di-n-octyl &0her: was prepared
as before using 2.8804 g (11 mmoles) triphenylphosphine, 1.8 g (11 mmoles) Br and 60 CH CN. When the mixture had warmed 2 mL 3 (10.0 to ambient temperature 2.4215 g mmoles) di-n-octyl ether were added in one portion. The flask was immediately fitted for simple distillation and the distillation work-up begun.
The fractional distillation yielded 0.57 g (3 mrnoles, 15% yield) 1.16 (4. 8 c 1-bromooctane and g mmoles) di-n-o tyl ether (48% recovery of the ether).
DTP I.A(3) Reaction of DTP with benzyl phenyl ether : was prepared as before using 60 mL CH CN, (20 rn,�oles) triphenylphosphine 3 5.2429 g and 20 mmoles Br After addition of the Br the mixture was 2 • 2 allowed to warm to room temperature and 1.8458 g (10.0 mmole s)
benzyl phenyl ether were added in one portion. The mixture was
heated to reflux (solution formed at reflux) and maintained there
The reaction flask was then allowed to cool , for 14.5 hours. 30 mL were added to dissolve the solid which formed and CH2c12 the entire reaction solution transferred to an addition funnel
attached to a 25 mL flask equipped for simple di stillation.
The CH CN and CH c1 were removed at atmospheric pressure 3 2 2 (Bp 39-83°) keeping the volu,�e of liquid in the flask at
approximately 15 mL until all of the reaction solution had been added to the flask. The flask was then set up for vacuum
di stillation and the material remaining in the flask was distilled 17
fractionally. The first fraction (a cle�r·colorless liquid with ° a small amount of white solid) was collected at 35-40 /22 mm ° (0.8502 g), the second fraction at 89-90 /17 mm (0.4125 g) and ° the third fraction at 91-92 /17 mm (0.0210 g). The liquid from fraction #1 was isolated by means of a transfer pipet, giving
0.4346 of clear colorless liquid. The three fractions g #1,
#2 and #3 were similar in appearance (clear, colorless) and each had a bitter odor (lachrymator!) similar to benzyl bromide.
Each fraction was injected independently on the gas chromatograph producing 3 chromatograms with similar shaped peaks having identical retention times. Co-injection of the three samples resulted in a single uniformly shaped peak with the same retention time as before. Co-injection of a sample from fraction #2 with reagent benzyl bromide (Aldrich) also produced a single uniformly shaped peak with a retention time identical to the previous chromatograms. An NMR spectrum of fraction #2 was identical to that of the Aldrich material. Fractions #1, #2 and #3 accounted for 51% of the benzyl bromide expected from complete cleavage of the ether. Three solutions were prepared: (A) a 10 wt% solution of benzyl phenyl ether, a 1.0 wt% solution of (B) benzyl phenyl ether and (�) a solution of the distillation residue, all.in The three solutions were spotted on CH2c12• a TLC slide (Analtech pre-coated x 100 mm with Silica Gel 25
at 250 microns). Solutions and each produced single GF (A) (B)
and R = a solution spots (Rf= o.589 f o.582Jrespectively) using 18
of 1 :1 n-hexane/toluene as the solvent. .Another TLC was prepared
comparing phenol (Mallinkrodt ), triphenylphosphine oxide and
the distillation residue. These 2 TLC slides are shown below.
The solvent system for the 2nd TLC was :1 CH C /CH CN. The spot from 1 2 2 3
1 ) wt% benzyl 1) phenol "' 10 - phenyl ether in 2) I.A(3} dist. residue
CH CL2 Ph P=o 2 �@ 3) 3 2) • wt% benzyl 1 0 I phenyl ether in
CH2c12 I1�e 3) I.A(3) dist. residue in CH2c12 -�
1 :1 n-hexane/toluene - . • 2 •I- • • • 1 3 I). 3 1 2
solution large and showed up very dark under irradiation (A) was UV
(the s�ides were pre-treated with a fluorescent) while the spot
from solution was smaller was fainter under'UV light, (B) and No spot was visible in that region or The only f solution (C). spot from solution (C) very point appeared near the spotting compared phenol (R = 0.099) to (Rr"" 0.093). A second TLC using f 19
more polar olv�nt (CH c1 /cH CN) show1' the presence of phenol. a �- 2 2 3 Injection of solutions (A), (B) and (C) individually on ° the GC under the same conditions (SE-30, 150 ) confirmed that no benzyl phenyl ether remained in the distillation residue.
Chromatograms of (A) and (B) exhibited reproducible peaks for benzyl phenyl ether. No pe for the ether was observed the ak in chromatogram of (C). A peak was observed on the chromatogram of (C) with a retention time identical to that of phenol. Co injection of solution (C) with the solution of phenol produced a single uniformly shaped peak wi th a retention time identical to the reagent phenol. Analysis of the distillation residue by GC and TLC methods showed the presence of phenol and the absence of benzyl phenyl ether. No phenol was isolated from the residue.
The white solid from fraction (0.4156 g decomposed #1 ) immediately upon addition of H2 o, evolving a white acidic gas . The solid produced by the decomposition was recrystallized from ° to give 0.274 g of white powder (MP 156-7 ). CH3 CN An infrared spectrum of the white powder was identical to that of some pre�iously prepared triphenylphosphine oxide. The solid from fraction #1 was unreacted DTP that had sublimed over during the distillation since of the crude after hydrolysis � a TL C material showed that no phenol was present. This eliminated the possibility of the white solid being the phenoxyphosphonium compound. 20
Reaction of with benzyl pheny-1.ether: was I.A(4) DTP DTP prepared as before using S.2423 g (20.0 mmoles) triphenyl phosphine, 3 2 g (20 mmoles) Br and 60 "When the mixture . 2 mL CH3CN. had wanned to ambient temperature 1.8543 g (1 0.1 mmoles) benzyl phenyl ether were added in one portion. The mixture was allowed , to stir at room temperature (200 ) for 18 hours. Work-up gave
0.59 g benzyl bromide (3.4 rmnoles, 34 yield) and 0.34 g % (3.6 mmoles, 36% yield) of phenol. and analysis of the TLC GC di stillation residue showed no trace of the starting ether.
I.A (S) Reaction of DTP with benzyl phenyl ether: Benzyl phenyl ether (1.8465 g, '1 0.0 mmoles) was treated with 20 rmnoles of DTP in 60 mL at reflux for hours. Benzyl bromide was obtained CH3CN 2 in a 74% yield (1.26 g, mmoles) and phenol was obtained in 7.4 a 49% yield (0.456 g, 4.85 mmoles). Analysis of the distillation residue by TLC and showed no trace of the starting ether. GC
I.A(6) Reaction of ether: Benzyl phenyl DTP with benzyl phenyl ether (1 .8400 g, 10.0 mmoles) was treated with 20 rmnoles DTP ° in 60 mL CN at room temperature (20 ) for 4 hours. Benzyl CH3 bromide was obtained in yield (0.73 g, 4. 3 mmoles). a 43% TLC and analysis of the distillation residue showed no traces GC of the starting ether. Phenol was found present in the residue but was not isolated. 21
. I.A(7) Reaction of DTP with benzyl phenyi ether: Benzyl phenyl ether (1.8395 g, 10.0 mmoles) was treated with 11 mm.oles DTP in 60 CH CN. Upon addition of the ether to the mixture mL 3 DTP the flask was immediately fitted for distillation and work-up was begun. Benzyl bromide was obtained in a 72% yield ( 1.23 g, 7.2 mmoles) and phenol was obtained in a 6Cffo yield (0.55 g, 6.0 mmoles).
TLC and GC analysis of the distillation residue showed no trace of the starting ether.
Preparation I.B(3) of triphenylphosphine oxide: DTP was prepared using 5.24 g (20 nnnoles) triphenylphosphine and 3.2 g
(20 mmoles) Br in 60 mL CH CN. After preparation the reaction 2 3 DTP mixture was diluted with 3 times the volume of H o. A milky 2 white oil formed which produced rhombic crystals upon cooling in ice water. The solid was isolated by filtration, dissolved in CH c1 , and the solution dried over Mgso • The solvent was 2 2 4 removed under vacuum and the solid slurried in 15 mL n-hexane with heating \'I/hen the hexane was warm, CH c1 was added dropvtlse . 2 2 until the solid had dissolved. The solution was filtered hot through fluted filter paper and �llowed to cool. White crystals
needles were isolated by formed ( ) which filtration to give a ° 156-7 first crnp of pure triphenylphosphine 0xide (1.8462 g, MP ).
The filtrate was reduced to one half of the original volume to ° 156-7 ). produce a second crop of white crystals (0.2945 g, MP
ine was The yield of triphenylphosph oxide 2.1407 g (38.6%) of 22
white powder #6235, IR #9529). (NMR The FMR and IR spectra of 2 2 the product were identical to literature spectra� ' 4
Standardization of cleavage products and recovered ethers for
reaction I.F(2): Four pre-weighed 10�00 mL volumetric flasks were used to prepare (1) a solution of 1.5053 g ethyl salicylate
in CH c1 , 2 ( ) a solution of 1.5017 g benzyl bromide and 2 2 1.5048 g 1-bromooctane in CH c1 , (3) a solution of 1.4969 g phenol and 2 2 1.4324 g benzyl phenyl ether in CH c1 and 4)( a solution of 2 2 1.4868 g di-n-octyl ether plus 1.508 g ethyl salicylate in CH2c12• (4) were From solutions ( 1 ) , 2), ( 3( ) and five :''Standard samples prepared by pipetting various amounts of the solutions into pre-
weighed sa�ple vials. These six standards were injected individually
on the GC using the 5% UCON SOLB 550 on Chromasorb W column. (A) The conditions were as follows: solutions I-II at column temperature 100°, attenuation 2, flow rate 60 mL/min, chart speed
1 , in/min , _detector current 150 rnA detector temperature 240°, 1 injector temperature210°, sample volume rnicroliter and (B) 11 ° solutions III-Vat column temperature 0 for 5 minutes then
150°� attenuation 1 , flow 60 mL/min , chart 1 in/min, detector
, current 150 rnA detector temperature and injector temperature sarne as (A), -and sample volume 1 microliter. All subsequent
chromatograms were made under the same conditions varying only
column temperatures, attenuation and sample injection volumes.
The analysis was performed according to procedures in the Results
section. 23
I.F (2) Reaction of DTP with benzyl phen:el · ether and di-n-octyl ether simultaneously : DTP was prepared as before using 0.2420 g
(0. 9 226 rmnoles triphenylphosphi ne, 0.1 6 g {1mmole ) ) Br2 and 60 mL CH3CN. When the reaction mixture had warmed to ambient 2.4285 {10.02 1.8400 temperature g mmoles ) di-n-octyl ether and g (9.988 mmoles ) benzyl phenyl 'ether were added simu ltaneously . ° The mixture was heated to 40 and kept there for 2 hours . The
0.5 mL ) reaction was then quenched by addition of (28 mmoles H o. Mgso /K co 2 The reaction so lution was dried over 4 2 3 then 64-82°) reduced in volume by atmospheric-distillation ( bp to approximately mL 1.0034 ethyl 15 and g salicylate were added . Using ethyl salicylate as the internal standard the reaction mixture was an alyzed quantitatively for products and recovered ethers .
Standardization of cleavage products and recover ed ether from reaction I.G(2): Three solutions were prepared in pre-weighed volumetric flasks: (1) 3 . 771 4 g of ethyl salicylate in 25 .00 mL
CH c1 solution, (2) 10.00 mL CH c1 2 2 1.5092 g diphenyl ether in 2 2 so and 1.4973 10.00 mL CH c1 lution (3) g bromobenzene in 2 2 solution.
Standard samples were prepared by pipe tting various volumes of
(1 ), w e so lutions (2) and (3) into pre- eigh d sample vi als. The . standard samples were injected on the GC, peak areas of the components determined and the detector response factor obtained . 24
I.G (2) Reaction of with diphenyl ether Eastman : Di.phenyl DTP ( ) ether (1 .6955 g, 9.97 mmoles was treated with mmoles of ) 11.0 in 60 mL CH CN at 40° for 2 hours. fter 2 hours the reaction DI'P 3 A was quenched by addition of mL H o. Upon addition of the 2.0 2
H20 a white crystalline solid which had been suspended in the mixture disappeared immedi ately. The solution was dried over
Mgso /K co , filtered, the filter cake washed with CH c1 and 4 2 3 2 2 the bulk of the solvent removed by atmospheric distillation
bp 50-83° ) until the volume of the liquid in the distillation ( flask was approximately mL. The solution was weighed and 15 two weighed aliquots removed for GC analysis : I.G(2)-I with
0.0300 g ethyl salicylate in o.6097 g of solution and I.G(2)-II with 0. 1 049 g ethyl salicylate in o.6034 g of solution. Samples of both aliquots were inj ected on the GC under the following conditions: I.G(2)-I column temperature 100°, attenuation 1
maximum sensitivity and sample size 1-3 microliters; I.G(2 )-II ( ) column temperature 150°, attenuation 4, injections #1 and #2 at 1 microliter and injection #3 at 1.5 microliters. Sample
I.G(2 )-I was intended for analysis of the cleavage possible products bromobenzene and phenol. No measurable peaks were observed for either of these two compounds. Sample I.G( 2)-II was used for ·quantitative analysis of the remaining diphenyl ether . The results of the analysis are shown in the following section. 25
D(5) I. Reaction of DTP with di-n-octyl ether : Di-n-octyl ether (2 .4236 10.01 mmoles was treated with 11.0 mmoles of g; ) DTP ° in 60 mL at 0 for hours. The reaction was quenched CH CN 4 2 3 by addition of mL the white solid present dissolved 2.0 H o ( 2 immediately , the solution was dried over filtered, Mgso ;K co , ) 4 2 3 the filter cake was washed with the bulk of the solvent CH c1 and 2 2 removed atmospherically bp 53-83° ) until the volume of liquid ( in the flask was approximately 15 mL. The solution was weighed and two weighed aliquots were withdrawn for analysis after GC addition of the standard: I.D (5)-I with g solution o.8042 containing 0.01 96 g ethyl salicylate and I.D(5)-II with o.8779 g solution containing 0. 11 64 g ethyl salicylate. The 50L 5% UCON B column was used for the analysis under conditions described GC for reaction Aliquot I.D(5)-I was used for analysis of I.G(2). 1 -bromooctane and I.D(5)-II was used for analysis of the recovered di- n-octyl ether.
I.A(8) Reaction of with b nzyl phenyl ether : Benzyl phenyl DTP e
ether (1 . 8421 10.0 rnmoles ) was treat e with 11. 1 rnmoles of g, d
DTP·in 60 mL CH CN at 40° fo hours. The reaction was quenched 3 r 2
with . 0 mL H 0 the white solid disappeared), the solution 2 2 ( dried
an d the bulk of the solvent removed atmospherically over Mgso4/K2co3 (bp 65-82° ) until the volume of solution roxi atel mL . was app m y 15 o i was weighed a o The s lut on weighed and two li qu ts were wi thdrawn
for GC analysis after addition of the standard : I.A(8 )-I with 26
.0742 g ethyl salicylate in 0.5984 g solution and I.A(8)-II with
0.0803 1.000 g ethyl salicylate in g of aliquot. Aliquot I.A(8 )-I was used for analysis of benzyl bromide in the reaction solution and I.A(8)-II was used for analysis of the recovered ether.
Standardization of cleavage products and recovered ether for reaction I.H(1 ): Four solutions were prepared in pre-weighed
10.00 1.51 37 g cyclohexanol (standard ) mL volumetric flasks: (1 ) in 10.00 CH c1 solution, ( 2) 1.5029 1-bromobutane in mL 2 2 g 10.00 mL solution, ( 3) 1.4361 g butyl phenyl ether in CH2c12 10.00 mL solution and 0.9993 g 2-hexanone (standard ) CH2c12 (4) in 10.00 mL CH c1 solution . Standard samples were prepared as 2 2 before using 2-hexanone as the standard for 1-bromobutane and cyclohexanol as the standard for butyl phenyl ether. Peak areas were determined by disc integration and the response factors obtained from the data.
1 Butyl phenyl I.H( ) Reaction of JJI'P with butyl phenyl ether : (10 .0 .50 6 11.0 ether mmoles, 1 1 g) was treated with mmoles of JJI'P· in 60 mL CH CN at 40° for two hours. The reaction was quenched 3 with 2.0 mL H o and the solution dried and reduced in volume as 2 before. Two· fractions were obtained in the removal of the solvent;
#1 35-81 °) 81 -85°). an s fraction (bp and fraction #2 (bp VPC aly is of fraction showed that of the unreacted #2 approximately 2 0% ether was recovered here. Two aliquots were re'!loved from the 27
reaction solution andwe ighed: with 0.01 52 g 2-hexanone I.H(1 )-I in 0.6476 solution and with .1336 g cyclohexanol in g I.H(1 )-II
0.7021 g solution. Ethyl salicylate was not a suitable standard for this reaction because of its long retention time on the GC compared to 1 -bromobutane and butyl phenyl ether. Several injections were made of each 'aliquot on the 5% UCON 50LB column under the following conditions : , )-I column temperature I.H(1 60°, flow 60 mL/min , attenuation 1 , chart 1 in/min, and ° I.H(1 )-II at column temperature 100 , flow 60 mL/min , attenuation 4 , chart in/min . In addition, two other solutions 1 (A)' 0.075 were prepared: Solution g 1 -bromobutane in 15 mL CH CN and Solution 1 -bromobutane mL CH CN. 3 (B) o.ol S g in 1S 3 Solution (A) represented a 5% (.55 mmole) yield of 1 -bromobutane
(based on 1 0 mmoles of starting ether ) and solution (B) represented a (.1 1 mmole) yield of 1 bromobutane, both in 1% - volumes of solvent approximately equal to the volume of the actual reaction solution. GC analysis of the reaction solution showed a verry sm all peak at a retention time equal to reagent 1-bromobutane. Comparison of this peak with injections of solutions and under identical conditions indicated t at (A) (B) p only 1-2% (.1-.2 mmole) 1-bromobutane had been produced in the reaction.
of va e Standardization clea g products and recovered ether from
: Three were reaction I.J(1 ) solutions prepared in pre-weighed 2 8
volumetric flasks : g phenyl cylcohexyl ether in 10.00 (A) 1.5094 solution, 1.5068 g bromocyclohexane in 10.00 rnL mL CH2c12 (B) s0lution 1.4937 g ethyl salicylate in 10.00 CH2c12 and (C) mL H Cl solution. Four standard samples were prepared as before C 2 for each component br oc cl hexane phenyl cyclohexyl ( om y o and ether ) and these samples were inject d on the GC using the e 5% column . Column temperatures were 0 for the brorno- UCON 50LB 100 ° cyclohexane and 150 for the phenyl cyclohexyl ether.
I.J(1 ) Reaction of DI'P with phenyl cyclohexyl ether : Phenyl cyclohexyl g, 9.90 rnrnoles was treated with 11.0 ether (1 .5076 ) ° mm les of in 60 rnL at 40 for 2 hours. The reaction o DI'P CH3CN was quenched with 2. 0 the white solid present at this mL H2o ( time disappeared , the solution was dried over and ) :Mgso4/K2co3 reduced in volume to about mL 2 2 g by atmospheric 15 (1 .131 ) distillation bp One aliquot (I. J(1 g ) ( 52-83°). )-r ·, o.6616 was withdrawn and g ethyl salicylate were added to it. 0. 0919 This aliquot was used to analyze the reaction solution for unreacted phenyl cyclohexyl ether. No products were observed for. this reaction .
Standardization of cleavage products and recovered ester from reaction : Two solutions in I.K(1 ) were prepared pre-weighed volumetric .5064 in 10.00 flasks : (A) 1 g ethyl salicylate mL 10.00 CH2 c12 solution and (B) 1 .S052 g benzy1 benzoate in mL 29
CH c1 solution. Standard samples were prepared as before and 2 2 injected on the 5% UCON 50LB collunn at a column temperature of ° ° 130 for minutes, then 180 for 8 minutes. The areas of the 5 peaks of the standard and the benzyl benz oate were obtained
then analyzed and the using a disc integrator. The data were detector response factor was · determined.
I.K(1 ) Reaction of DI'P with benzyl benzoate : Benzyl benzoate
DI'P (2.1380 g, 10.1 mmoles) was treated with 11.0 nnnoles of ° in 60 CH CN at 40 for 2 hours. The reaction was quenched mL 3 by addition of H o, dried over Mgso /K co and the volume 2.0 mL 2 4 2 3 of the liquid reduced to ab out 1 g . Aliquot �O mL ( 24.75 7 ) I.K(1 )-I (0 .7005 g) was withdrawn and 0.01 62 g ethyl salicylate were added. The aliquot was then analyzed by GC using the column conditions described in the previous segment.
II.A ( 1) DI'P was prepared as before mmoles ) in Blank : (11.0 ° 60 mL CH CN . The mixture was heat d to 40 with stirring and 3 e maintained at that temperature for 2 hours. The reaction was qu enched h 2.0 H o (the white solid present di sappeared wit mL 2 the solution dried over Mgso /K co and reduced immediately), 4 2 3 in volume about GC analysis of reaction solution to· 15 mL. the showed only the presence of CH CN and CHc1 (the latter had 3 2 2 reaction washing of Mg so /K co been added to the solution by the 4 2 3 or drying agent ·with CH2c12). Some slight irregularities ttbum:is" 30
were observed on the baseline under highest instrument sensitivity, but these were so small that their identification could not be determined. Contact of Ul'P hwit CH CN under these 3 conditions produced no significant amount of products and the two materials can be considered to be non-reacting in relation to one another.
Extraction of asphaltol (toluene-insoluble pyridine-soluble) fraction from Coal PSOC 252 (from Penn State Coal Data Bank ), an Illinois No . 5 coal : A Norton 2.5 L ball mill was flushed N 120 252. with 2 and charged with g of coal PSOC The coal was initially ground for 45 minutes using 8 ball-grinders (Sargent 27.9 Welch, g , corundum ) then for another 30 minutes using 16 ball-grinders. The ground coal was sieved using wire mesh screens
1 125 a (ASTM 20 mesh, f'm)and shaker apparatus used by geologists for screening rock and mineral samples (a setting of 7 for 30 minutes was used). The portion of the coal passing through the
120 mesh screen was collected. This coal (98.5 was placed in g ) a 3.0 liter 3-neck round-bottom flask equipped with a mechanical N stirrer, condenser and 2 inlet tube. The flask was charged with 1 .5 L pyridine (MCB ). Some gas wasevol ved when the pyridine was added to the coal . The mixture was stirred at room temperature for 18 hours, then heated to reflux for 2 hours. The mixture was allowed to cool and then filtered (Whatman No. 1 qualitative filter papers ). Centrifugation of the filtrate (8200 rpm for 2 hours) showed no sediment. The filtrate was stripped of pyridine 31
to yield a tarry solid whi ch was stored in· the dark under vacuum . The recovered coal was returned to the 3 liter flask
1 . and wa s extracted with another 5 L pyridine. Thi s procedure
( extraction, filtration and stripping of filtrate on the rotary
evaporator ) wa s performed 5 times. The final filtrate was clear and the color of strongly brewed tea. The previous filtrates had been solid black in color. All of the extracts were combined, reduced in volume to about 100 mL and stored under vacuum as a vi scous black liquid.
Pyridine from the stripping of the filtrates wa s dried over
KOH for 3-4 hours wi th stirring , then distilled from BaO for use in further extrac tions. After the fifth extraction and filtration
th e recovered coal was slurried a final time in cold pyridine
( room temperature ) for 1 hour then filtered. Thi s pyridine washing was combined wi th the extracts a". The coal was filtered, wa shed with deionized water, air dried then placed in an
Ab derhalden apparatus and dri ed under vacuum (S0°/o.5 mm ) to
constant weight . The final weight of the extracted coal was 71 .5 g whi ch was 7 2 .6 % of the original coal •
. The concentrated extracts were poured wi th stirring into
10 times the volume (1 liter ) of toluene and the mixture wa s allowed
to stir for h5 minute s . The mixture was filtered and the solid
wh ich remained in the fil ter wa s washed with toluene, 1 Cffo HCl
and H o. Thi s solid wa s the asphaltol fraction . Th e asphaltol 2 32
t Abderhalden (100°;0.2 was dried to constant weight in he mm). A total of 1 3.1 277 g of black, medium grain asphaltol was obtained (13.3% yield) . This asphaltol wa s designated 252-B. The re toluene/pyridine filtrate was duced in volume to about 10 50 mL then poured into a times volume of n-hexane with 45 s stirring . The mixture was stirred for minute , filtered and {asphaltene) was the solid washed with hexane, 10%HCl and H2o. air a total of 3.4 g t ai d Upon drying of bro'Wll ma erial rem ne (3.5% yield) . This was desiginated 252-C.
hexane/toluene a e The filtr t was reduced in volume to a ht green thick, viscous oil which fluoresced a brig in the light much like petroleum derived oil. This oil was desiginated 252-D and about (2.5% yield). weighed 2.5 g
Samples of all of the coals and coal fractions ( excluding the b t in the asphaltene and oil ) were com us ed a muffle furnace to determine the ash content. The method of ash determination
ASTM D 31 74 5 used was the standard method� This procedure is en h described fully on page 33. Elem tal analyses of t e whole extracted 252-B btain from coal, the coal and were o ed Galbraith Labs · {Knoxville, Tenn. ) . The molecular weight of the asphaltol was determined to be 1040�
The method used to determine the molecular weight of the
ph t t , a h s Vapor Pres sure as al ol (and la er the reacted sp altol ) was Osmometry (VPO). The procedure required that a carefully weighed 33
sample of the asphaltol be di ssolved in pyridine (the solvent used for the a haltol dete tions ) and then inj ected i to sp MWn rmina n
e designed oven. 2 j t o por s a sp cially This oven contained in ec i n t one which were placed such that each allowed inj ection of a liquid sample e s . The ar e onto a th nni tor oven was ch g d with pyridine and
to a t o ere heated moderate temperature (while sealed) until he atm sph
e t e within th oven was sa urated with pyridin vapors. When equilibrium was ch eved, 2 liquid samples were injected simultaneously a i pure e into the oven : ( 1) A sample of pyridin (reference ) and (2 )
di The change in a sample of the asphaltol dissolved in pyri ne . vapor pressure experienced by the thermistor injected with the asphaltol/pyridine solution compared to that of the thennistor
p was re or ed inj ected with pure yridine c d . The instrument had been pre-calibrated using samples of known molecular weight so
particular v or pr�s sure was r l that a change in ap di ect y attributed
the number of l e . For exa�ple : to partic es causing the chang If
pr e v a 0. 1 g sample of asphaltol oduc d a apor pressure change
, equivalent to 0.001 mole of particles then the number average molecular weight (MW .) of that sample would be 100. n
standard method of ash determination in samples of coal and coal fr ct o 31 74 : This method of a i ns, ASTM D ash determination required the following items :
(1 ) Porcelain crucible
c and/or (2) Dessicator ontaining CaC1 Dririte 2 34
Crucible tongs (3 ) .. #30939 ) (4) Muffie furnace (in thi s case a Hotpack model, .Ea.stern
The crucible was cleaned by heating a solution of o.5 mL cone . HCl and cone . HN0 the crucible over a low flame 3 mL 3 in until re h bro was evolved. After cooling, the crucible ddi s - wn N02 , was rinsed several times with deionized water . The crucible was
w then dried to constant eight over a Me eker burner (usually 2 hours eatin was ) h g sufficient and stored in a dessicator until
e needed. After drying, the crucible was handled only with th crucible tongs.
A sample of the co or coal fraction was into the al ( ) weighed cruc ble sample ) . then dry i (100-200 mg The crucible was placed in the cold muffle furnace and the thermostat set at 30 such
a ur that temperature of 5oo0o was achieved in 1 ho and a temperature 2 s of 150°0 in hour . When the temperature had reached t s e 750°, the thermostat set ing wa chang d to 25 to keep the furnace temperature stable at 750°. The sample was heated at
total of 3 in 150° for 1 hour ( a hours the furnace from start finish , then the crucible was removed placed on a ceramic to ) and n e plate to cool for about 2 minutes. After 2 mi ut s, the crucible was placed in the dessicator and allowed to cool at least 1 hour before �ighing . After weighing, the crucible was cleaned and dried as before . 35
R o DTP III.A(1 ) eacti n of with asphaltol f52-B : A 0.501 0 g
(0.48 mmole) sample of asphaltol 252-B was treated with 11. 0 mmoles of' DTP in CH CN. The r eaction was carried out 60 mL 3 in a 100 mL 3-neck round-bottom flask equipped with a magnetic stirrer, heating oil bath, condenser and N inlet tube. The DTP was prepared 2 as before. When the reaction ' mixture had warmed to ambient temperature, the asphaltol was added in one portion to the reaction flask. Slight vacuum was applied to the flask to degas the asphaltol. When the asphaltol was sufficiently "wetted" in thi s manner, the flow was resumed to the flask and the mixture was N2 ° heated to 40 and maintained at that temperature for 115 hours. by The reaction was quenched addition of 2.0 mL H o. The stirring 2 was discontinued and the solid allowed to settle. The super- natant liquid was clea r brown in color. The mixture was filtered
(Whatman #1 qualitative fi lter paper), the solid was washed with 1 (III.A(1 and then slurried in CH c1 (about 00 mL). The solid )-a) 2 2 was dried to constant weight (1oo0;o.25 mm) to give 0.51 27 g of black material (102% yield from starting asphaltol, NMR # 621 8, IR # 9814).
·The solubility of III.A(1 )-a in was determined by c:r:x:a3 g 1.0 placing 0.050 of the material in mL of CJ)(Jl3• A�ter g The filtr�te filtration, 0�032 of the material was recovered. was dark black in color and centrifugation of the filtrate
(8200 rpm for 5 minutes) produced no sediment. The solubility
of III.A(1 )-a in was about mg/mL • When a mg sample crxa3 18 10 36
9Cffo of III.A{1 )-a was placed in 1.0 mL pyridine, approximately of the material appeared to be insoluble. The infrared spectrum 1 of III.A{1 )-a showed strong absorptions near 1 150 cm- characteristic
of triphenylphosphine oxide or possibly a phosphonium compound.
Elemental analysis showed 1 1.4% Br and 1.82% P. III .A{1 )-a
(95.7 mg)was treated with 50'mL of 10%HC1 /CH CN (a 1:1 volume 3 2 solution ) for 4 hours at room temperature. After filtration and 71 .6 drying to constant weight, mg of material remained. The (197.9 remainder of III.A(1 )-a mg)was treated in the same manner.
.• After drying, the acid-treated material (III.A(1 )-h) weighed HiC.3 !J}.g This was a weight loss of 49 .6 mg (25%). Urr The 148.3 mg of III.A(1 )-h was sl ied for 1 hour in room temperature pyridine. The mixture was filtered and the clear brown filtrate poured into a 10 times volume of 10%HCl {aqueous ) . The HCl/pyridine mixture was stirred for 1 hour and filtered.
The solid 'rtas. washed -with 10%HC �, H2o anQ. ace�,one, �hen drj_ed to constant weight in the Abderhalden to yield 15.4 mg cf pyridine
soluble solid (IR # 9666 ).
2 2 A III.A(2) Ex:traction of asphaltol 5 -B with acetonitrile:
3-neck round-bottom flask was charged with 60 mL CH CN 1 00 mL 3 2 and 0.5026 g (0.48 mrnole) of 25 -B. The asphaltol was wetted by application of a slight vacuum to the flask . The flask was then heated under r (with stirring ) to 4o0c and kept at that for 115 to temperature hours. Water (2.0 mL)was added the mixture 37
and the flask was allowed to cool to ambient temperature. The mixture was filtered the solid washed with The solid and CH3CN. (III.A(2 )-a) ( material was dried to constant weight 0.4�25 g, IR # 9668 ). A total of 0.0801 g of material (15.9% ) was lost during the interaction the When- 50.1 with CH3CN. mg , of III.A( ) - a was slurried in 1.0 for 10 minutes at 2 mL cnc13 26.2 room temperature then filtered, mg of material were recovered as insoluble.
The first filtrate ( of the reaction mixture) was dried over ·
filtered and reduced in volume on the rotary Mgso4/K2co3, evaporator. This produced a thick viscous material (0. 0667 g, III.A(2 )-b ) which was similar in appearance and odor to the oil · fractton obtained from the toluene/hexane extract of the coal.
III.A(4) Reaction of with asphaltol 252-B: o.5002 g DTP A (0.48 nnnole) sample of 252-B was treated with nnnoles of 11.o ° in 60 H N for 11.5 hours at 40 c. The reaction was run IYI'P mL C 3C using the same procedure as for reaction III.A(1 ). After 1 1.5 hours, the reaction was quenched by addition of 2.0 H o. mL 2 The inixture was allowed to cool, was filtered and the insoluble material (III .A(4)-a) was washed with then dried to CH2c12, 8 (98.1%, constant weight. The yield of III.A(4)-a was 0.490 g IR # 9815). The solubility of III.A(4)-a in pyridine was determined
0 sample in o.5 of pyridine. After by slurrying a 1 .1 mg mL filtration (Whatman filter papers), there was a #1 qualitative 38
slight amount of residue on the filter paper and the liquid
of was clear. Elemental analysis III .A(4)-a showed 9.08% Br
1.42% The was determined to be 508 by VPO. and P. MWn .
III.A(4)-a (2 4 . 6 mg)was treated with 50 10%HC1/CH CN 7 mL 3 for 24 hours (with stirring) at room temperature. The mixture was ltfi ered, the solid washed with HCl and H 0, then 1 0% 2 dried to constant weight (0.21 88 g, IR # 9669 ). This was a 21 % decrease in weight. The HCl treated material was designated III.A(4)-h.
The molecular weight of III.A(4)-h was found to be 2239 by VPO.
e and to Th Br P percentages had gone down 6.32% and .24%, respectively. 39
Results and Discussion ...
The cleavage of ethers reported by Anderson and Freenor 21 ° (p. 10) was carried out in benzonitrile 122-30 ) and DMF (@ ° 60-1 0° ). Both of the solvents are high boiling (bp 191 (@ 1 for benzonitrile and bp 1.53° for and are difficult to remove DMF) from products boiling in the same range. The change to acetonitrile as the reaction solvent logically called for longer reaction times to offset the lower reaction temperatures.
The first ether studied was di-n-octyl ether. Reaction ° times of 16 and hours (at reflux, 81 ) gave 43% (8.6 mmoles) 17 and (9.0 mmoles ) yields of 1-bromooctane along with 18% (1 .8 4.5% rmnoles ) and 8% (0.8 mmoles) recovery of the starting ether, respectively. Cleavage of the ether was occurring but isolation of the products and unreacted ether was not complete. There was the possibility that the small amounts of water present even in the "dry" acetonitrile (which had been di stilled from CaH) were hydrolyzing the dibromotriphenylphosphorane (DTP) before cleavage could take place. For that reason the quantity of DTP in the reaction was increased from a 2� mole excess to a 10� mole excess.
This.did not seem to affect the amount of cleavage significantly.
The only product observed in the cleavage reactions of di-n-octyl ether was 1-bromooctane.
It was also possibile that almost complete cleavage was occurring but one half of the 1-bromooctane was bound up in the form of the phosphoniu.i� ion (Equation 11, p. 11 ) intermediate. 40
t It was found in later experiments that addi ion of aqueous HCl or H20 to the reaction mixture at the end of the reaction gave better recovery of products from the reaction .
The next model ether enzyl phenyl ether. After hours was b 1 7 of reaction with DTP a 51 % (5.1 mmoles ) yield of benzyl bromide was obtained. Analysis of th distillation residue showed the e presence of phenol, but no starting ether remained . Un der the same reaction conditions and approximately the same reaction
l times; IJrP had produced 45% c eavage in a dialkyl ether and what
pp e v e nzyl- l a eared to be 1 00% cl a ag in a be ary ether. Thi s was
selective cleaving g . the first hint that DTP could be a rea ent In order to test this further, the reaction time of DTP with benzyl phenyl ether was reduced to 2 hours at reflux with a yield of mrnoles) benzyl bromide and a yield 74% (7.4 49% (4.9 mrnoles) of phenol. Analysis of the distillation residue showed that no starting ether was present.
The next step was to reduce the amount of DTP in the reaction to a mole excess and to decrease the reaction time so that 10% the distillation work-up was begun as soon as the ether had been
s added to the prepared DTP. Thi allowed for a contact time of the ether with the of 'about minutes to hour. Reactions DTP 45 1 di-n-octy1· zy p ny (separately ) of ether and ben l he l ether under these conditions yielded 15% (3.0 mmoles) 1-bromooctane and 73%
mrnoles) benzyl bromide. Cleavage of the benzyl phenyl ether (7.3 produced a 60% (6.0 mrnoles ) yield of phenol. No benzyl phenyl 41
ether was found in the distillation residue for the benzyl phenyl ether/DTP reaction and 48% (4.8 mmoles di-n-octyl ether were ) of recovered from the di-n-octyl ether/DI ? reaction. DTP was obviously showing selectivity for the benzyl-aryl ether over the dialkyl ether. This seems reasonable that a benzyl ether would in be more reactive toward a reagent like DTP than an alkyl ether. Reactions of DTP with benzyl phenyl ether at room temperature
were inconclusive, since a 34% mmoles yield of benzyl (20° ) (3.4 ) bromide was obtained after 18 hours and a 43% (4.J mmoles yield ) of benzyl bromide was obtained after only 4 hours. All of these results are summarized in Table 2.
In thi s series of cleavage reactions, isolation of the products and artingst ether was a problem . Much of thi s was due to the nature of the reaction mixture during the work-up. Depending on the amount of DTP used in these reactions anywhere from 3-5.5 g of solid material (triphenylphosphine oxide and unreacted DTP) remained in the final reaction mixture. This relatively large amount of solid hindered efforts to isolate small amounts of liquid products and ethers after the reaction.
·Once it was known th at DTP would cleave benzyl phenyl ether even at room temperature the second phase of the study pegan;
low temperature reactions of DTP with model ethers. The 1£: ies Instead work-up of this second ser of reactions was modified. of distilling the reaction mixture for physical isolation of the products and unreacted ethers, the reactions were analyzed Table 2 Ether Cleavage Reactions at 81° in Acetonitrile Using Dibromotriphenylphosphorane (IJI'P)
# Mrnoles Mmoles RBr # Mmoles Other Mmoles Ether Rxn # Rxn RBr Ether Ether IJI'P Time Product Yield �%� Products (%) Recovered (%l # 2 43%) 10 1 16 h 1'1-C H 8 .6 ( 1.8 (18) 1 ( C8H17)20 8 1 .7Br ..----- 5 2 (C H, 0 0 20 17 h n-c8H1 �r 9.0 (4 ) - o.8 (8) 8 7)2 1 -.--- - b H C H 0 10 1 di still n-C8 1 Br 3.0 (15) 3 ( 8 17)2 1 7 ------4.8 (48) 2 5., ( 51 PhOH a 4 PhCH20Ph 1 0 0 17 h PhCH2Br ) O.O (O) a 2 ' PhOH o.o (O) 5 PhCH20Ph 1 0 0 2 h PhCH2Br 7 .4 ( 74 ) (49 ) .i::- b a I\) 6 ' distill PhOH o.o 9 PhCH20Ph 10 1 1 PhCH2Br 7.3 (73 ) (60) ( )
° Ether Cleavege Reactions at 20 in Acetonitrile Using Dibromotriphenylphosphorane (IJI'P) •
PhOH a 7 PhCH20Ph 10 20 18 h PhCH2Br 3.4 (34) (36) O.O (O) a 8 PhCH20Ph 10 20 4h PhCH2Br 4.3 (43 ) PhOH O.O (O)
{ ) VPC a Analysis of dis�illation residue by TLC and showed less than 1% starting ether b Distillation was begun at addition of ether to IJI'P; reaction about to ( ) time 45 min 1 h 43
quantitatively by VPC using an internal standard. The analytical procedure was as follows : Solutions were prepared containing known weights of the compound to be analyzed in the reaction mixture and another compound chosen as the standard for comparison . purposes. The weight ratio of the compound !. to the standard s s 3 - (w.1 /ws ) was lmown for each tandard solution. Usually to 5 solutions were prepared with the weight of the !. component varying from about one half that of the standard to about one and one half that of the standard to give a good point spread in the calibration. Samples of each of the solutions were injected on the gas chromatograph and the areas of the peaks produced by the compound interest (A. ) and the standard (A ) were determined of 1 s by Disc integration. The ratio<:of the weights of the two materials was proportional to the ratio of their GC peak areas by the proportionality constant � called the detector response factor.
In some cases the detector response was linear (or nearly so ) and � was simply the slope of the line defined by equation 13.
w./w • k (13) 1 s A1/A s
In other cases the response was not· linear and the data had to be treated using the_·relationship shown in equation 1 4.
( 14)
The data were analyzed using a library least squares routine on the MICC Computer System. In the first case (linear 44
response) the data were analyzed by a first-order least squares
and the second case (non-linear response) by a polynomial fit in second-order fit. This computer analysis of the data gave the constants and The data were then applied to analysis !f,1 !!, !!_2 • of the reaction mixtures. At the end of the reaction, a small amount of water was added to hydrolyze the unreacted DTP, the solution was dried, filtered and the bulk of the solvent was removed by atmospheric distillation. A weighed amount of standard was added to an aliquot (of known weight ) withdrawn from the pre-weighed reaction solution. Samples of the aliquot were inj ected on the GC, the peak areas of the standard and
and unknown component were determined by Disc integration, the amount of the unknown in the reaction solution was calculated using the detector response factor(s) for that particular compound. The reaction solutions were analyzed for expected products and starting ether remaining at the end of the reaction .
Only glass GC columns we re used in the analyses because it was found that benzyl bromide decomposed on metal columns .
The first reaction in the second phase of the study was the competition of benzyl phenyl ether (10.0 mmoles) and di-� o a small amount of DTP nnnoles) ctyl ether (10.0 mmoles ) for (0.92 at 40° for 2 hours (reaction The results and experimental I.F(2)). data for the standardization are shown in ble Ta 3. Graphs of
F w /w vs A./A for each component of are shown in igures i s 1 s I.F(2) and results of the GC analysis of the reaction solution 2 3. The are tabulated in Table There cleavage 4 mmoles 4. was about 48 % (.4 Table 3 Ex:perL�ental Results From Standardization of Reaction I .F( 2) Using Ethyl Salicylate As Internal
Standard Component A k k Smnple wi (g ) ws (g ) wi/ws i/As 1 2 --. + + IA PhCH2Br . 0?508 . 1 505 .4989 4167 1 . 31-.04 - .06-.01 IB PhCH Br .1502 • 1 505 . 9980 . 851 8 2 IC PhCH Br . 300 . 1505 1 .9953 2.0625 2 3 + IIA C Br . 09072 . 1 508 .601 6 .6450 . 99-.04 8H1 7 IIB C H Br . 1210 . 1 508 . 8024 .8484 a 1 7 IIC B 1 2 .1508 1.003 1 .030 CaH1 7 r . 5 1 .i::- IID Br . 1 514 . 1 508 1.203 1.1 90 \1\ c8H1 7 1 7 IIE c H1 7Br .2 1 . 1508 , .404 1.430 8 + . + IIIA PhC Ph . 505 .4759 . .5441 . 93- .02 -. 068-.005 H20 .07162 1 IIIB Ph 0Ph . 1 432 . 1505 .951 5 1 .1 5 CH2 1 IIIC Ph H 0Ph .2868 . 1505 1 .9056 2 .528 C 2 + + 8 4 - • o IVA C 0 .08922 . 1508 . 591 6 . 1050 . 8 0 &:.01 . ( 8H1 7 )2 -. IVB 0 • , 1 90 . 508 .7891 .8920 ( C8H1 7)2 1 C 0 . 1.21 0 IVG ( 8H1 7 )2 . 1 487 1 508 .9861 IVD ( C H 1 784 .1508 1.183 1 .380 8 1 7 ) 20 . IVE ( C H 0 . 2082 .1508 1 . 381 1 670 8 1 7 )2 . + VA PhOH .07484 . 1 505 .4973 .61 76 .1�.02 2 PhOH 50 .• 7 1. 50 VB .1497 .1 5 994 . 2994 . 1 505 1 .9894 2 vc PhOH .. 717 h6
Reaction Graph of w./w vs . A./A I.F(2) l s l s V From Tabl e and 3
2.0
1 . 5
r.1 /w s
1 . 0
.5
= 0 c8H1 7Br PhOH o=
.6 .8 1.0 i.2 1.41. 1 6•
A./J_ A s .
Figure 5 Reaction I.F (2) Graph of w./w vs . A./A For Samples III 1 s 1 s IV From Table and 3
2.0 .
1 .s
w.l /w s
1.0
. s
0 = PhCH2 0Ph
a= (C5H1 7)20
.h .6 .s 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0
A.l /A s Figure 4 48
Table 4 Quantitative Analysis of R a tion I.F(2 . Ethyl Salicylate e c ) Using Internal As Standard
Component A k k w ( ) mrnoles Ai s Ai/As 1 2 s g w1 (g,calc. ) 1 PhC Br 1 488 .o645 1 31 -.06 , .0034 . 0841 .492 � 3 . 5 . .0540 , .31 1.0034 .0105 .41 2 42.0 777 - . o6 24. o 455 . 05 27 1 .31 - . 06 1 . 0 034 .0688 .402 1 7.5 3o6 .0 2 1.31 - . 06 1.0034 ' .0747; , .437 57 49.0 806 . 0608 , 1 .31 - . 06 , .0034 . 0793 .464 + . -+ Ave : .01&:.02 44 .03 PhOH 1a. o 306 .0588 . 10 1 . 0034 .041 3 .439 42.0 806 .0521 . 10 1.0034 .0366 .389 56. o 1008 .0556 . 10 1.0034 .0390 .41 5 24.0 488 .0492 . 10 1 .0034 .0346 . 368 40.0 777 .051 5 . 10 , . 0034 .0362 .385 + + Ave : .039- .002 .4�.02
CH 0 488 1.93 .93 - .01 1 . 0034 1 . 55 8 42 Ph 2 Ph 944 . 1 530 777 1 . 97 .93 - .01 1 . 0034 1 .57 B .53 960 455 2.10 .93 - .01 1 . 0034 1.66 9 . 02 605 306 1 .97 . 93 - .01 , .0034 1 .58 B .59 1 6 56 806 2.05 .93 - .07 1.0034 1 . 63 8.86 1 936 1008 1 .92 .93 - .01 1.0034 1 . 54 B.31 + ·+ Ave : 1.59-.04 8.6- .2
2.2 488 .00451 . 85 1 . 0 034 .003B5 . 0199 CaH1 7Br 4.5 777 . 00579 . 85 1.0034 . 00494 . 0256 4.5 806 .oo55B . 85 , .0034 .00476 . 0247 1008 . 00595 . 85 1 . 0034 .00507 .0263 6.o + Ave : . 0047-+ . 0004 . 024-.002
H ) 81 3 - 306 2.66 . BB 1 . 0034 1.93 7.97 (C8 17 20 - . 06 1 . 0034 2 .01 8.30 2263 806 2 .81 .BB -.06 .BB , . 0 034 2.02 8.35 1 31 2 488 2.69 -.06 2202 777 2.83 . 88 - . 06 1 . 0034 2.00 8.26 + Ave : . + 0 1 98-. 4 8.2-.2 49
benzyl bromide produced) of the benzyl phenyl ether and only about
cleavage mmoles 1 -bromooctane produced) of the di-n 1.3% (0.24 octyl ether. In this manner it was sho'W?l that DTP would selectively cleave a benzyl-aryl ether in the presence of a dialkyl ether under mild conditions. ' Treatment of 1 o.0 mmoles benzyl phenyl ether with 11 mmoles of Ul'P under the same conditions produced mmoles (80%)be nzyl 8.0 bromide with about 1.h mmoles ( 1h%) benzyl phenyl ether remaining after 2 hours (reaction I.A(8)). The results of the GC analysis are sho'W?l in Table 5.
Treatment of 1 0.0 mmoles di-n-octyl ether with 11.o nnnoles ° of DTP for 2 hours at h0 produced 1.6 nnnoles 1 -bromooctane (8%) with analysis showing almost 99% (9 .9 mmoles ) of the ether remaining at the end of the reaction (reaction I.D(5 )). The results of the
GC analysis for reaction I.D(5) are shown in Table 6. When mmoles of diphenyl ether were treated with 11.0 9.97 mmoles of under the same conditions as the previous two DTP reactions, no cleavage was observed (reaction I.G(2)) and 8.8 mmoles ( 88% ) of the diphenyl ether was recovered. The standardization data for diphenyl ether are shown Table The results of in 7 . the GC analysis are shown in Table No products were observed 8. from this reaction including phenol and bromobenzene.
Butyl phenyl ether (10. 0 mmoles) was treated with 11. 0 ° mmoles of under the same conditions as before (40 for 2 DTP hours ) to produce about 0.9 mmoles 1-bromobutane (9 .0%) with 50
Table 5 Quantitative .Analysis of Reaction I.A(8 ) Using Ethyl Salicylate
As Internal Standard
Component A A A. /A k k w (g, calc.) mmoles i s J. s 1 2 ws (g) 1 1 FbCH2Br 65.0 89.0 .730 1 . 31 -.06 . 0742 1.32 . 7.72 , 21 1 59 . 759 1 . 31 -.06 .0742 1 . 37 8 .01 , 1 5 152 . 156 1 . 31 - . 06 .0742 1.36 7.95
6 90. 0 .850 1.31 - . .07 2 1.52 8.89 7 .5 06 4 63 .0 86. 5 .728 1 . 31 - .06 . 0742 1 . 31 7.66 54. o 72.0 . 750 , . 31 - . 06 .0742 1 .31 7 .89 + + Ave : , . 37-.05 8.0..:. . 3
PhOH 76.o 1 59 .5 .476 . 70 .0742 .475 5 . 05 64.0 1 52 .421 . 10 .0742 .420 4.47 37.5 90.0 .417 . 70 .0742 .416 4.42 37 . 0 86.5 .428 . 10 . 0742 .425 4. 53 36.5 12 . 0 .507 . 10 . 0742 .506 5.38
46. o 56.o . 821 • 70 .0803 . 531 5.65 66. o 85.o .776 . 10 .0803 .502 5.34 98.5 1 27 . 776 . 10 . 0803 .502 5. 34 + + Ave : . 47- .06 5. o..:. .5
- PhCH 0Ph 18. o 90.0 .200 .93 .01 .0742 .261 1 .42 2 1 5. 0 86 . 5 . 1 12 . 93 - .01 . 0742 . 226 1.23 1 4.0 12 . 0 .194 .93 - . 01 .0742 .253 1.38 42.0 127 . 332 . 93 - . 07 .0803 .211 1 • .50 25.o 85.o . 294 .93 - .01 .0803 . 247 1.34 22.0 56.o . 393 . 93 -. 07 . 0803 .328 1.78 + + Ave : 26- 02 . . . 1 4-.1 51
Table 6
Quantitative Analysis of Reaction I.D(5) Using Ethyl Salicylate As Internal Standard
4
Compo ent A. A A /A k k w w. g, calc . mmole s n i s 1 s (g) 1 1 s 2 1 ( ) C3H17Br 29 .0 42.0 .690 .99 .01 96 . 321 1 . 54
57 .0 79.0 • 721 .99 . 01 96 .335 1 .74 12. 0 1 7 . 5 .686 . 99 .01 96 . 31 9 1.65 26.0 39.0 .667 .99 .01 96 .31 0 , .61 + + Ave : . 32- . 01 1.66 .: .04 C 0 57.0 56.o , .018 .9 -.05 . 01 96 . 4 , o., ( 3H,7 )2 3 2 5 69. 0 71 . 0 . 972 . 93 - .05 .01 9 6 2.35 9.7 1 56. 5 56.o , . 009 .93 - . 0 5 .01 96 2. 43 1 0.0 + + Ave: 2.41- .04 9. 9-.2
Table 7
Experimental Results of Standardization of Reaction I.G ( 2 ) sin U g Ethyl Salicylate As Internal Standard
Component A A A./A k w (g) w w./w s i s i s (g ) l s 1 + . 0 .1508 .6004 49. 5 88. 5 . 55 . 91-.02 PhOPh 09 54 9 • • 1 0 .8004 28.0 34.5 .811 , 207 5 8
• 0 .1509 1 508 . , .001 32.0 32 . 0 1.0
• 1 1508 1 .201 36. o 29.0 1 .24 , 81 . . 21 1 3 . 1 508 , . 401 39 .0 27.0 1 .44
Table 8 of e c o Using Ethyl Salicylate Quantitative Analysis R a ti n I.G(2) As Internal Standard Component A A./A k k w (g) w. g,calc . mmoles. A. 1 2 1 s 1 s s l ( ) l 45.0 .662 . 9 1 • 1 049 1 .42 8.35 PhOPh 68.o 46.5 64. 0 . 726 .91 • 1 049 , . 55 9 .12
. 1 • 1.52 58.5 82.0 9 , 8.94 .71 3 049 + + Ave : 8.8- . 3 1.50-.05 Table 9 . Experimental l s of standardization of Reaction I.H(1) Usin Resu t g
Cyclohexanol (Standard for Butyl Hlenyl 2-Hexanone Ether) And Standard fo ( r 1 -Bromobutane)
Component (g ) (g A A J., /A k wi ws ) wi/ws i s i s + . 09018 . 0999 . 024 52.5 85.5 .61 4 1.44-.03 c4�Br 3 9 ' . 1202 2. 3 .865 .09993 1 .203 3 0 7.0 .1503 . 099 3 1.504 33. 31 .5 . 048 9 0 1
+ H10C • 46.o 4� .0861 6 1 51 4 . 5961 26.0 .565 .94- .03 • 1149 • 151 4 . 7589 29 5 40. 5 28 . . 7 • 1 436 • 1 514 .9485 32.5 35. 0 . 928 . 1 723 ' .151 4 1 . 1 38 36.5 33. 0 1.1o6 .201 0 . 1514 1.328 4o.o 29.5 1.356
,· '
Table 10
Quantitative Analysis of Reaction I .H(1 ) Using Cyclohexanol And 2-Hexanone ·As Standards Internal
Component A. A A./A k k w (g ) w. (g,calc. ) mrnoles 1 s 1 s 1 2 s 1 1 c4�Br 1.5 52. 0 .0288 1.44 .01 52 . 0116 .078 2.0 6. . 0 .01 52 1 3 • 5 5 354 .0 4 104 1 .44 + + Ave : .01 J..:. .001 .09 -.01
& 4o. o ·79.0 . 506 . 94 . 1336 1.08 1.1 9· oc4� 42.0 82. 5 . 94 • 1 336 1 09 7.27 S06 . 25.0 49. 5 . 505 .94 . 1 336 1.08 1. 19 + + Ave : 1 .08-.01 7 .22-.04 of the ether recovered at ,the end of the reaction (I.H( 1 )). 94% The experimental results for the standardization are listed in
Table and the results of the quantitative analysis are in Table 9 10. About 20% ( 2.0 mrnoles of the butyl phenyl ether was recovered ) in the removal of the solvent. Another mmoles was 72% (7.2 )
accounted for in the reaction' mixture at the end of the reaction .
The last model ether tested was phenyl cyclohexyl ether.
When mmoles of this ether were treated with 11.0 mrnoles 8.�5 at 40° for 2 hours no cleavage was observed reaction of DTP ( I.J{ 1 )). GC analysis of the reaction showed no products were
present mmoles of the starting ether remained. and S'!�� ( 8 � 06 ) The standardization data are li sted in Table 1 1 and the results
of the GC analysis are listed in Table 12.
Table 11
Experimental Results of Standardization of Reaction I .J( 1 Using ) Ethyl Salicylate As Internal Standard
Component w. w w /w A. A A./A k 1 s i s 1 s 1 s + Bromo- .09041 • 1494 .6051 23 .0 28.5 .597 1.03- .04 cyclohexane .1205 • 1 494 .8066 53 .5 72.0 .743
.1507 • 1494 , .009 35.0 34 . 5 , .01 4 .1808 . 1 494 1.21 0 37.5 32 .5 1.154 + enyl- .0?0.56 • .6062 106 .623 89 08 Ph 1 494 66.o . -. cyclohexyl .1207 • 1 494 .8079 6J.O ao.o .788 Ether . 1509 .1494 1.01 0 79.5 Bo.a 1 .oo6 • 81 1 • 1 494 1 . 21 2 73.5 1 . 292 , 95.o Table 12 Quantitative .Analysis of Reaction I. J(1 ) Using Ethyl Salicylate As Internal Standard
Component A./A k (g,calc. mmoles A A 1 ws g wi 1 i s s ( ) ) . 9 1.38 Fhenyl- 102 11 .923 . 8 9 091 l. 8---3' cyclohexyl o.5 1 06 , 1 4 .934 .89 .0919 1.40 Ether 7.94 48.o 49 . .9 0 .89 .0919 1.46 5 7 8;2.a 87 .0 92 .0 .946 .89 .09 1 9 1.42 a.-06 + - Ave : 1 .42-. 02 + a.0.:..1
The last reaction with a model compound involved attempted
cleavage of an ester, benzyl benzoate (10. 1 mrnoles), with 11.0
mrnoles of IYrP. After 2 hours at 40° a 3% cleavage (0.3 mrnoles of benzyl bromide produced of the ester was observed reaction ) ( I.K(1)). The data from the standardization are listed in in Table 13. The results of the GC analysis are shown Table 14.
About 105% ( 10.6 mrnoles) of the unreacted ester was accounted
for.
Table 13 · Experimental Results of Standardization of Reaction I.K(1 ) Using Ethyl Salicylate As Internal Standard
w w./w k Component w g s A.1 /A i s 1 s li A s (g) ( ) + PhCH 0 CPh .499 61 .0 91.0 1 .6-.3 2 2 .07526 .1506 6 .610 64.0 1 .024 .1505 .1506 .9992 62.5 199 . .1806 .1.506 1. 10 0 6.5.o 1.077 Table 14 ) i y at Quantitative Analysis of Reaction I.K(1 Using Ethyl Sa.l c l e As Internal Standard
Component A A A k ( g ) (g,calc. ) mmoles. i As w w i/ s s i 1 PhCH2Br 5.0 10.5 .0709 1.31 . 01 62 . 0530 .31 0 65.0 .o692 , . 31 .01 62 .051 7 02 4.5 3 + . + Ave : . 5 -. o 06- 00 o 24 o o6 .3 . 4
PhCH202CPh 128 51 . 0 2.51 1.6 .01 62 2.30 10. 8 192 79 . 0 2.43 1.6 . 01 62 2.22 10.5 + Ave : 2.26:.04 1o.6!. 2
The results of the low temperature cleavage reactions of
DTP with model compo ds are tab la ed in Table 1 The un u t 5. greatest percentage of cleavage occurred with the benzyl phenyl
ether (reaction #1 , Table to the extent of cle age 15) 80% av
nnnoles of benzyl bromide produc ) econd highest (8.o ed . The s percentage of cleavag e was with di-n-octyl ether ( reaction
#2, Table 1 5 with 8% clea age ( 1 .6 nnnoles of 1-bromooctane ) v produced). The most probable mechanism for the reaction of
dibromotriphenylphosphorane with benzyl phenyl ether and
di-n-octyl ether is somewhat different than the mechani sm
21 proposed by derson and Freenor F.quations (10)- 1 2 ), p . 11 ) . ki ( ( 6 The Denny a1� showed that t r i phosphine work of � e t ary polar solvents s ch as existed in dihalides in u CH3CN ,o
equilibrium with the tertiary phosphine cation and the halide anion (Equation (15)). It is this tertiary phosphine bromide . halide species which probably iniates the cleavage of the ether
( 1 5) +
Equations ( 1 6) and illustrate this mechanism with enzyl (17) b phenyl ether and di-n-octyl ether, respectively •
• • �Ph + r (16) PhCH20-Ph + :lB ----� -Br
PhOH C C�H2��- 8H17 + Ph Br -·---� Br + (1 7 ) J� Ph P+ Br- 3�� ':>7
F.quations and It can be seen from {16) (17) that reaction
of DTP with an alkyl ether goes to completion, while reaction of with DTP a phenyl ether stops at formation of the bromotriphenyl
phenoxyphosphine. This moiety (16a) is stable thermally (decomposes
above 230°C) and requires addi tion of water to yield the final produc�s of phenol and triphenylphosphine oxide.
Reactions of DTP with the other model compounds (reactions #5 mmole) #3, #4, and #6) all yielded less than 3% (0.3 of cleavage products. Since the cleavage is believed to occur by
SN2 attack, the inertness of diphenyl ether was expected. The
order· of r&activities'to 8rJ2 (or �1 ) attack is as follows : 20 Benzyl ) octyl > > phenyl. Burton and Koppes found that
cleavage of esters of halogenated acids occurred much more readily
than cleavage of esters of non-halogenated acids. This would
explain the low percentage of cleavage of benzyl benzoate.
In the competitive reaction between equimolar amounts of
benzyl phenyl ether and di-n-octyl ether with a small amount of
DTP (#7 ), the benzyl phenyl ether showed about a 37 :1 relative
reaction rate over di-n-octyl ether. The high reactivity of
benzyl phenyl ether with DTP is another indication that these
reactions occur primarily .by S 2 attack. In comparison reactions, N benzyl halides are generally at least a hundred times as reactive 27 as ethyl halides in displacement reactions. SN2 !>0
Table 1 5 Reactions of Dibromotriphenylphosphorane (DTP) With Model Compounds 2 at 0 in Acetonitrile For Hours 4o c Mmoles Mmoles Yield Compd. % %ec vere # Co!!!Eound Com:ed. IYI'P RBr RBr R o d 11. 0 8 . 1 PhCH20Ph 1 0.0 PhCH2Br 0 1 4 1 11. r 99 2 (C5H1 7 >20 0.0 0 n-C8H1 7B 8 1 ---- 3 PhOPh 9.97 1 .0 -- - 0 88 4 1 1 2 Phoc4� 0.0 11. 0 c4�Br 9 8.85 1 1 . - -- -- 4 5 PhOC6H1 1 0 - - 0 9 6 PhC0 CH Ph 10. 1 11. 0 PhCH Br 106 2 2 2 3 * PhCH20Ph 1 0.0 PhCH2Br 48 86 7 0.92 2 (C H ) 0 1 0.0 n-C Br 1.3 * 8 8 1 7 2 8H, 7 * Based on starting DTP
The work of using dibromotriphenylphosphorane to cleave model
compounds was now complete. The results show that the reaction
most likely occurs by coordination of phosphorus to the ether
oxygen, then S 2 displacement on the alkoxy carbon by bromide ion. N Benzyl phenyl ether wa.s the most reactive towards IYI'Pof the
compounds studied. Cleavage of benzyl phenyl ether was complete
n ether remained at the end the reaction ) at (!! o starting of ° 0 81 c, and was 80% complete at 4o c. It was now time to turn to tol fraction Illinois reactions of DTP with the asphal of an No. 5 coal, The pyridine-soluble toluene-insoluble- fraction (asphaltol ) used in these experiments was extracted from Illinois No . 5 coal
(PSOC 252 from the Penn State Coal Data Bank). The asphaltol fraction (252-B ) was 13.5% of the original coal by weight. The ultimate analyses of the whole coal, the extracted coal and
252-B are shown in Table 16. 'The number average molecular weight of 252-B was found to be 1040. The asphaltol was totally soluble in pyridine (a 50 mg sample dissolved in 2.0 mL pyridine to give a clear dar.k brown solution which left no residue when filtered through Whatman #1 ��alitative filter papers and which showed no
for n ) . sediment after centrifugation at 8200 rpm 10 mi . The material was insoluble in CH c1 , CHC1 and acetone. When 10 2 2 3 mg of 252-B in 1.0 of a solvent (� CH c1 ) produced a clear mL 2 2 colorless liquid after agitation for 5 minutes, then filtration, the material was said to be insoluble in that solvent.
Table 16
Ultimate Analyses of Whole Coal (Illinois No. 5) And Extracted Coal Fractions (dry ash free basis, weight percent )
Sample c H N (diff) Ash O&S MWn Whole Coal Penn State 79.57 . · 5.42 1.76 , 3.12 12.59 -- -- figures ('79) Analysis (after 72 . 74 5.47 , .58 20.22 11. 90 grinding, t81 )
Extracted Coal 74. 01 5.23 2.03 18. 73 10.40
Asphaltol 76.67 5.65 2.28 1 5.39 2 . 38 1 040 252-B 60
- An infrared spectrum of 252-B is shoWn in Figures 5a&b . 8 The prominent absorption bands are assigned in Table 17� The 1 absorptions in the region of 1050-1 300 cm- are indic ative of phenols, alcohols and ethers.
Table 17 28 Prominent Ab sorption Bands�of Asphaltol 252-B -1 Frequency ( cm ) Assignment 3200 0-H stretch 3050 arom. C-H stretch 2900, 2850 aliph. C-H stretch ' 1 600 arom. C•C stretch1 H-bonded y=O 1450 aliph. -CH - and -CH 2 3 1300-1000 l C-0 phenols, Ar�O-Ar, C-0 a cohols �0-��� 870, 81 4,760 arom. out of plane C-H bending
Two samples of 252-B were treated with DTP at 40° in acetonitrile : One sample for 115 hours (reaction III.A(1 )) and
the other sample for 11.5 hours (reaction III.A(4)). Two such
different reaction times were chosen in order to compare long-tenn
and short-tenn reactions of DTP with 252-B. The solid CH CN 3 - insoluble portion from each reaction was isolated. When 1 0 mg
of the product from the 115 hour reaction (III.A(1 )-a) were_ placed
, in 1.0 mL pyridine there was a large amount of insoluble material.
ima io the l i i y -a pyri n An est t n of so ub l t of III.A(1 ) in di e would c
PS � 1601 cm-1 3500 3000 2500 2000 1500 1 20( 4000 1 (Frequency cm- )
Figure 5a
-IR Spectrum of Asphaltol 252-B c "
PS 102 8 cm -1
7 0 1 360 1 2�0 1 1 cio 1 o�o 9bo a'oo � 6�o 5�0 4�< 1 (Frequency cm- ) Figure 5b IR 252-B Spectrum of Asphaltol be e 1 mg/mL. When 10 mg of the product from the 11.5 hour r action
( III.A(4)-a) were placed in . 5 pyridine, all but a small amount 0 mL a was soluble. An estimation of the solubility of III .A(4)- in pyridine would be about 1 8 mg/mL. The molecular weight of
. III.A(1)-a could not be determined because of its low.solubility
in pyridine. The of III.A -a was found to be 508, a 51% MWn (4 ) decrease. Ultimate analyses of both III.A 1 -a and III.A(4)-a ( )
are compared with the original asphaltol in Table 18.
Table 1 8
Ultimate Analyses of Asphaltols After Reaction With Dibromotriphenyl- ° Phosphorane (Dl'P) in Acetonitrile at 40 (daf, weight percent )
Rxn c Br Sample Time H N oecs (cliff ) p H/C MW III .A(1 )-a 1 1 5 h 35 4.82 1.53 9.97 1 .81 11.33 .816 10. III.A(4)-a 1 1 . 5 h 72.58 5., 72 1 .42 9 . 1 2 .847 508 6 1. 10.00 76.67 5�65 2.28 15.39 .896 1040 252-B
As Table 18 shows, both III .A(1)-� and III .A(4)-a have
significant percentages of P and Br. The ratio Br/P was 2.40 for III.A 1 )-a and 2.46 for III.A(4)-a. If we assume that all (
of the P was in the form of either unreacted DTP or a phosphonium complex, we could correct the Dl'P-reacted asphaltol analyses for
the C, H, Br and P contained in those materials (the unreacted or P honium complex . this manner, it is possible Dl'P hosp ) In to compare the reacted asphaltols quantitatively between
5 themselves and the original 2 2-B. If we assume that all
the P is in the form of DTP, we are also assuming that no cleavage took place. The 51 % decrease in the of III.A 4 )-a MWn ( after 11. 5 hours reaction wit DTP, however, strongly implies h that cleavage did occur. Therefore, we assumed that the P was + in the form of a phosphonium complex (R-O-PPh Br ) such as would 3 be formed if cleavage took place. Table 19 shows the corrected
analyses.
Table 19
Phosphorus Corrected Analyses of DTP-Reacted Asphaltols Along ¥li.th the Original 252-B
Rxn % Yield % c Sam le Time Based on #1 c H diff Br Rec . H/C ;e N O&S ( ) (1 ) 2$2-B 76. 67 5.65 2.28 15. 39 .878 -a 115 h 102.3 72. 1 5 4.92 1.91 1 4 . 60 8.32 77. 0 .81 3 (2) III.A(1 ) (3) III.A(4)-a 11. 5 h 102.6 74. 35 5.30 2 .04 1 1.86 6 .45 83. 8 .847
From the results in T ble 1 9, one can see that after 115 hours � reaction time _with DTP 23% of the orig na C was lost, after i l while 11.5 hours reaction time with DTP 16. 2% of the original C was
lost. The H/C ratio of both reacted samples is lower than in
the original asphaltol, as is the 0 S content The Il/C ratio and . of the DTP-reacted asphaltols could have been lowered by the loss of some aliphatic material from the coal structure, or by loss of and in the appropriate amounts or by loss of H H2o co2 by oxidation. The oxygen and sulfur content could have been lowered
cleavage of ether linkages and loss of oxygen to triphenyl by phosphine oxide or by loss of and One must bear in mind, 'H2o co2 • however, that the percent of oxygen and sulfur in the analysis is the least accurately determined, incurring as it does the combined uncertainties of the other elements. Nevertheless, the presence of bromine in the DTP-reacted asphaltols beyond that expected to be associated with a phosphonium complex is additional evidence that some type of cleavage occurred.
When one compares infrared spectra of the DTP-reacted asphaltols {Fig. 6a&b, 7a&b ) with that of 252-B, several differences are discovered. Samples and both show III.A(1)-a III.A(4)-a � 1 -1 1 decreased absorption in the 2900 �m- , 1440 cm and 814-tl70 cm- regions of the spectrum. These regions represent, respectively, aliphatic stretching, -CH - (scissoring) and {rocking) C-H 2 -CH3 and aromatic groups (bending ). The decrease vibrations -CH- in the absorption in the first two of those regions indicates that aliphatic carbon and hydrogen have been lost during the reaction with DTP. The elemental analysis data supports the spectral data on that point. Decrease in the latter region 1 {81 4-870 cm- ) suggests the loss of aromatic rings, or further � PS 1601 cm·1
< I I I ( I I I I 4000 3500 3000 2500 2000 1500 1 201 (Frequency cm-1 )
Figure 6a
IR Spectrum of III.A(1 )-a � PS 1 028 cm-l
700 1300 1200 1 100 1000 900 800 600 500 40 - (Frequency cm 1)
Figure 6b
IR Spectrum of III .A( 1)-a � PS 1 601 cm-1
I I I I I I I 4000 3500 3000 2500 2000 1500 , 200
(Frequency cm-1 ) Figure 7a
IR Spectrum of III .A(4 )-a c �
PS 102 8 cm- 1
1300 1200 1 1 00 1000 900 800 700 600 500 4Q( 1 (Frequency cm- )
Figure 7b
of IR Spectrum III.A(4)-a -,u
substitution of the aromatic rings present. Another difference -1 1 e spectrum of in the spectra occurs at about 040 cm • In th IR 252-B a small absorption band is observed in thi s region, which is generally a ssigned to C . -0-C . and C h -0-C carbon- a 1ip aliph arom alip h oxygen stretching. This band is not present in the IR spectra of
III.A(1)-a and III.A(4 )-a. T e final obvious differences in the h -1 spectra are the , 740-690 and ab sorptions at 1120 cm cm-1 , 1 540 cm· • These absorptions are similar to those exhibited by the spectra of the triphenylphosphine oxide .These absorptions indicated the presence of the phosphonium complex and possibly some triphenylphosphine oxide.
An spectrum of th CDC1 -soluble portion of III.A(1 )-a NMR e 3 w as obtained in the following manner: A 50 mg sample of III.A(1)-a was placed 1.0 mL crx::1 in a sample vial (capped under N ), in 3 2 shaken vigorously then allowed to sit for 24 hours in a warm ° (25 ) place. Filtration of the mixture later recovered about
The filtrate was transferred to 32 mg of the solid material. tube. filtrate color. an NMR The was clear amber in The NMR spectrum of this material shows a multiplet of peaks at 7.5 ppm
(Fi . g 8) which are very similar to the multiplet observed in the
NMR spectrum of triphenylphosphine oxide. Farther upfield in
r · 1 • 6 and • . the egion of 1 • 3 ppm are two smaller multiplets These� peaks represent the aliphatic protons in the asphaltol structure.
The weak multiplet at o.6 ppm is assigned to a spinning side�band 9 of These assignments follow those of Bro-.,,m et al • ·what TMS . � I'd I ,..., .-'4 ...__, < H H H co 4-1 Q) 0 H
.61..... µ. µ� (,) Q) p.. 0 U) 0 M � �
0 ·o .,
0 -o "' ''
appears to have happened is that the cnc sol 13 ubilized some
of the small er phosphonium complex-aspha.ltol fragments. The spectral amplitude on the T 60 - NMR spectrometer necessary to obtain the spectrum was 800. The phenyl protons from the phosphonium complex (or phosphine oxide ) mask the aromatic region, so we can gain no further :information from it. Neither 252-B nor III .A(4)-a were soluble enough in to obtain comparison cnc13
NMR spectra of them.
In an effort to hydrolyze the suspected phosphonium complex, both III.A(1)-a and III.A(4)-a were treated with a 1:1 volume solu ion of aq. for 24 hours at 20°c . The resulting : 10% HC1/CH3CN products, III .A( 1)-h and III .A(4)-h, represented recovery of
67.3% and 63 .6% of the original carbon (based on 252-B),respectively.
Since III .A(1 )-a represented a loss of 23.0% of the original c, then III .A(1 )-h represented a loss of another of the original 13.4% (based on 252-B). III.A(4)-a represented a 16.2% loss of C original and III.A(4)-h represented another 20.2% loss of the C original c. Table 20 shows the combined phosphorus corrected results of the elemental analyses of the DTP-reacted asphaltols, the hydrolyzed asphaltols and 252-B. The values in parentheses are the values reported by Galbraith and the Univ. of Illinois .
As one can see from Table 20, P and Br percentages for both
DTP-reacted asphaltols are lower after hydrolysis with 1 0% HC1/CH3CN.
These data support the assumption that the bulk of the P was in "(j
Table 20
Phosph rus Corrected Analyses of DTP-Reacted Asphaltols, y e o H drolyz d Asphaltols and the Original 252-B Orig . c Orig . Sam;ele % % H c H N O&S ( diff) Br p Ash Recovered Recovered 252-B (76.67 ) (5.65)(2.28) 15. 39 2.38
III .A(1)-a 72.15 4.92 1 .91 14.60 8.32 -'!"- .60 77.0 71 .3 (70. 35 ) (4.82) ( 1.�3 ) (9.97) (11 .33 ) ( 1 .82 )
I .A( 1 ) -h 73.02 4.52 2.1 0 1 3 . 13 7.21 .60 67.3 56. 5 I I (72.46) (4.51 )(1.98) ( 1 2.38 ) (8.14} (.52 ) I a 83.8 1 II .A(4)- 74.35 5.30 2.04 11.86 6.45 .40 81 . (72.58 ) (5.1 6)(1. 72) (i o.oo) (9 . 1 2 ) ( 1 .43) II .A(4)-h 72.26 4.74 2.26 14. 74 5.99 .40 63.6 56.6 I (72.07) (4.73 )(2.21 ) (14.42) (6.32) (.24)
Valu y and University Illinois es in () are the values reported b Galbraith of Elemental Analys is Lab the form of a phosphonium complex. Infrared spectra of III.A( 1 )-h
(Fig. 9a&b ) and III.A(4)-h (Fig . 10a&b) also support this assumption . . . -1 Both s ect a show no absorption in the 1 2 6 0 cm Ar P=O ) and 1120 cm p r _, ( 3 (P-C stretch for tri-aryl phosphines) regions of the spectra along
540 -1 stretching e The with the cm region ( P-0 mod ) . spectra also s e -1 how d decreased absorption in the 740-690 cm region (mono-subst. a o t s . The a e at -1 r ma ic ) dis ppearanc of the band 1260 cm is indication that some triphenylphosphine oxide that had been trapped within the the hydro y s asphaltol structure had been removed during l si . The e e e an yse and the IR spectra ca e that the of l m ntal al s all indi t bulk e t reaction with was the P pr sen in the asphaltols after DTP removed reaction with by 1 0% HC1/CH3CN. -J +:""
PS 1601 cm _,
' 3500 3000 2500 2000 , 500 1200 4000 (Frequency cm-1 ) Figure 9a
IR Spectrum of III.A(1 )-h 75 0 0 -=t'
0 0 \.I'\
0 0 ['--
..c: 0 I 0 - a:> ...... -
<• 's 0 H H -8..CD H H � � Ct-i �Q) bO 0 0 0 °' &Q) � �� - - 0 it Q) 's 0 co �p:; (\J H 0 - 0 0 0 ..- � -
0 0 C\J .....
0 0 CT'\ ..... PS 1 601 \ / cm- i
______I • l I _L______J 3000 500 2000 500 20( 4000 3500 2 j , 1 (Frequency cm-1 ) Figure 10a
IR Spectrum of III.A(4)-h ...... , ...... ,
1 028 cm-1 v PS '•
. 1300 1 200 1 1 00 1000 900 800 700 600 500 400
(Frequency cm-1 ) Figure 10b
IR Spectrum of III.A(4)-h 78
L 10 1.0 When mg of III.A(1 )-h was stirred in mL pyridine (room temperature) for 5 minutes, a considerable amount of insoluble material remained. The filtrate (filtration through
Whatman #1 qualitative papers) was light brown/yellow in color.
Approximately 8 mg of III.A(1 )-h was recovered during filtration. 4 When 10 mg of III.A( )-h was stirred in 1.0 mL pyridine (room temperature ) for 5 minutes, only a very small amount of insolubles remained. The filtrate (filtration through
Whatman #1 qualitative papers) was clear dark brown. and
Approximately 0.5 mg of material was recovered during filtration. The nwnber average molecular weights of III.A(1)-h (pyridine soluble portion) and III.A(4)-h as determined by VPO are 3150 and 2239, respectively.
It was decided that we would investigate the changes (if any ) brought about in the asphaltol by long-term contact with
CH CN. When a o.5026 g (0.483 mmole ) sample of 252-B was stirred 3 0 in CH CN for 115 hours at 4o c, a 15.9% loss in weight 3 (60 mL) (0.0801 g) was observed. An IR spectrwn of the product (III.A(2 )-a) is shown in Figures 11a&b . The spectrwn shows decreased -1 absorption at 2850 cm and 1 450 cm ·• Both of these regions -1 are aliphatic regions, so this indicates that aliphatic carbon and hydro�en were lost during the interaction of asphaltol with
CH CN. 113 3 The molecular weight of III.A(2 )-a was found to be 1 . This also suggested that some low molecular weight material
(probably was lost s ol. The elemental aliphatic ) from the a phalt analysis of III.A(2 )-a is shown irt Table 21 . . PS 1801 .6 cm- 1 J
-J '°
I I I I I 1 3500 3000 2500 2000 15I00 1200 L.ooo -1 (Frequency cm )
Figure 11a
IR Spectrwn of III .A(2 )-a ()) 0
1028 _, PS cm
1300 , 200 , 100 1000 900 800 700 600 500 400 1 (Frequency cm- )
Figure 11a
IR Spectrum of III.A(2 )-a 01
The filtrate from reaction III.A(2) was reduced in volume
te to an oily ma rial (III.A(2)-b). Thisma terial was similar in appearance and odor to the oil fraction 252-D extracted from the
. 2 - (taken whole coal. An infrared spectrum of III A( ) b using the sample, hence the presence of the 01 ) is shown Figures m-m CD 3 in ' 2 in _, 1 a&b . This spectrum shows absorption the 3500 cm (strong, due somewhat to D o added to sample tube), 2900 cm-1, 2 NMR - cm 1, cm-1 and cm:1 regions. indicates 1600 1 450 cm-1 ,1 050 940 This
that most of the material in III.A(2)-b is probably aliphatic, with a small amount of aromatic material scattered throughout.
Table 21
Ultimate Analy8es of Asphaltol 252-B and III.A(2)-a (Acetonitrile Extracted 252-B) % Yield Sample c H N (cliff ) Ash Based on ( 1 ) O&S H/C MWft (1 ) 252-B 2.28 15.39 8 .878 1 4 76.67 5.65 2.3 0 0 . - 2.13 17.03 2 1 1 ( 2 )III A.(2) a 77.81 5.49 .40 .841 1 1 3 84.
(Dryash free basis)
In with the extraction of 252-B CH CN of the original 3 1 4.6% C (based 252-B ) and 1 8.2% of the original H on original was lost
• 1.09 ) •. The of III.A(2)-b (Fig. exhibits peaks (H/C NMR 1 3) at 2.0-2.9 1 .35 3.75 ppm, ppm, and o.85 ppm. The spectrum was 29 the solvent. Ladner have taken using CDC13 as Brown and assigned
th • t 2 "" b enzy ose 1 H regions J;· ::o al spec ra as f o 11 ows : -4 ppm 1'ic ; 1 • = ; o.4 - 1.1 = 1 - 2 ppm gem;:· 1.J. aliphatic ppm aliphatic -CH3 � 1 If PS 1 601 cm·
I l I I I I ' 1500 I hooo 3500 3000 2500 2000 1200 {Frequency cm·1 ) Figure 12a . IR Spectrum of III.A(2 )-b CX> \.tJ
-1 / PS 1028 cm �
(Frequency cm- t ) Figure 12b
IR Spectrum of III.A(2)-b 0
.0 I ,...,. ')' I <"I ' '-' ;. � < • 1-1 :i:i:: t-1 "' t') ...... :n r-4 :1; :11 \11 ti Cl> 0 ', 1-4
•rlg, r.:.. +.>� 0 Cl> p. ti) �
-, -...::.. ::-� --',.!;..: ."!- -... - .. --..:;�--;·· �=- •t;:_ ��·� ...... __.. :=;-: ;,� -4 - -�,_ The multiplet at 3.75 disappeared upon addition of to the n2o NMR sample, therefore this set of peaks was caused by exchangeable protons. The NMR data, along with the IR and elemental analysis data on III.A(2 )-b strongly suggest that the bulk of the material is aliphatic carbon and hydrogen whi ch has been solubilized by CH CN. the 3 It is probably aliphatic hydrocarbons or alcohols with a few aromatic hydrocarbons or phenols.
Now let us review our data. We know that : ° (a) When asphaltol 252-B was reacted with DTP at 40 c for 115 hours, the product (III.A(1 )-a) was pyridine insoluble and contained 77.0%of the original C and 71 .3% of the original
H (based on 252-B). Spectral data (IR and NMR) showed the presence of a phosphonium complex and a small amount of triphenylphosphine oxide within the asphaltol. Elemental analysis showed the presence of Br even above that expected to be associated with a phosphonium complex. ° (b) When 252-B was reacted with DTP at 40 c for 11. 5 hours, the product (Ill.A(4)-a) was pyridine soluble (except for a tiny amount of insolubles) and contained 83 .8% of the original C and
81 .1% of the original H (based on 252-B ). Spectral data again
·showed the presence of a possible phosphonium complex or residual triphenylphosphine oxide. Elemental analysis showed an amount of Br above that expected to be associated with a phosphonium complex. The molecular weight of the product was found to be 508. OU
(c) When both of the DTP-reacted asphaltols were treated with
HC1/CH CN, ;the resulting products contained only 67.3% 10% aq. 3
of the original C and 56.5% of the originnl H (for the 115 hour
product) and 63. 6% of the original C and 56.6% of the original H
(for the 1 5 hour reaction material) based on 252-B · for 1 . . The RWn
the pyridine soluble portion of the !�5 hour product was 3150 and
the for the 11.5 hour reaction product was 2239. Spectral data MW'n on both hydrolyzed DTP-reacted asphaltols showed no phosphonium
complex or triphenylphosphine oxide .
(d) When 252-B was extracted in CH3CN for 115 hours at 40°c the material recovered contained only 85.4% of the original C and 81. 8%
the original H based on 252-B. Spectral data d of (IR an NMR) indicated
that the CH3CN soluble portion was mostly aliphatic, with some
aromatic material included • h �n� of the CH CN-extracted T e n 3
asphaltol was 1113. appare t for the lost material is The n MWn 77rY_O There are several possible explanations for the increase in
MW o f the DTP-reacted asphaltols after h drol s i s with the 10% HCl/CH N· n y y f
(1) The material analyzed for the d t erminations was not totally . MWn e
soluble in pyridine . For exampl e : A 1.0 g sample of material is found
to cause a vapor pre ssure change equivalent to 0.01 mole of particles.
0.9 g of . t Only he material was actually soluble, however, s o the vapor n pressure cha ge attributed to 1.0 g of s amp l e was in reality caused by only 0.9 g of material. Thus , the mol ecular weight of the material
calculated (l .O er g/ 0. 01 mole= 100 g/mole) is high than the actual
= molecular weight (0.9 g soluble material/0.01 mole particles 90 g/molc) . This type of error may have occurred in the detenninations � MWn of the reacted-asphaltols because they were not totally soluble in pyridir.e. (2) Intramolecular reactions which occurred during or after the reaction of the DTP-reacted asphaltols with 1 Cffo HCl/
reSlilted in the formation of igh molecular weight molecules CH3CN h A in the coal structure. (3) large quantity of low molecular weight material was lost during the reactions of the asphaltols
1(JfoHC 1/CH CN. This with DTP then 3 would suggest that all but the largest aggregates had been removed from the a,sphaltol structure. (4) The coal fractions were oxidized. Oxidation of the coal fractions would remove H from the coal structure in the form of H2o and/or add o. The most likely explanation for the increase in of the MWn asphaltol samples is a combination of all of the above possible causes. The insolubility of the reacted asphaltols in pyridine makes it entirely possi�le that some of the increase was due to experimental error during the analysis, but the bulk of the increase must be attributed to a combination of the other 3 factors.
It would require an error of 200-JO<:Jfo to account for such a large
increase otherwi se. MWn Factor 2 has a great deal of data to support it. We can assume (with reasonable confidence ) that cleavage of the asphaltol samples by DTP did occur . II). the case wh ere reaction 11. 5 with DTP occurred over an hour period, 16.2%of the original C was lost (based on the starting asphaltol ) and the MWn of the DTP-reacted asphaltol was found to be 508.� ·This cleavage would have left a number of reactive functional groups behind it. For
example: A benzyl bro�de could be left from cleavage of a benzyl
ether, an acyl bromide could be left as a result of a reaction between DTP and a carboxylic acid group, a phenoxyphosphonium
complex from reaction with a phenol, �· It is possible that these functional groups were not in position to interact during the reaction with DTP or work-up and drying. Some of the functionalities
ether which has been cleaved to an alkyl may have been masked, � an or benzyl bromide and the alkoxyphosphonium complex . Since the
of the 11.5 hour DTP-reacted product was found to be 508, �n most likely the bulk of the intra�olecular reactions took place during the hydrolysis with 10%HC1/CH 3CN. An example of the type
of reaction which may have occurred to increase the numb er average molecular weight of the asphaltol samples is shown in Equation (18).
Suppose 2 adjacent layers of the asphaltol structure were left,
as a result of cleavage by DTP, with functional group s that could
react with one another under the right conditions.
2 DTP
- HBr In this manner, the 2 adjacent layers which may have been separable by solvent action before are now chemically joined, increasing
the average molecular weight of the asphaltol sample.
Factor 3 also has evidence in its favor. A large quantity of material was lost from each asphaltol sample during the reaction
. one instance almost one third with DI'P and the hydrolysis In of the original C and one half of the original H based on the originnl asphaltol was lost the hour reaction product . ( 115 ) study has shown that extraction of the asphaltol with CH CN Our 3 for this time period hours was sufficient to remove almost (115 ) of the original C and of the original H whi ch and 15% 18% ( NMR IR showed was predominantly aliphatic material . Thus, the ) increase in molecular weight most likely owes a great deal to factor 3.
Finally, factor 4 must be considered. Oxidation of the coal fractions cannot be ruled out entirely. Considerable care was
taken to keep the samples in an inert atmosphere without ( resorting to use. of a glovebox whenever possible. Momentary ) exposures. to the atmosphere occurred in transferring the coal
fractions from their storage places to reaction vessels. During - filtrations, washings, etc. the coal fractions may have come in
contact with the atmosphere. Last, but not least, when the
samples were sent off for analysis they were not under vacuum or N2• However, it seems that oxidation would tend more to break down the coal structure louer its rather (� MWn ) than 90
increase it.
Conclusions
It appears that treatment of the toluene-insoluble pyridine-soluble {asphaltol) fraction of an Illinois #5 coal
DTP has reduced (initially ) the molecular weight of one sample with to 508, a 51% decrease from the original of 1040. Under the MWn conditions of the reaction (40°c, 11. 5 hours ), the only significant cleavage observed in a study of model ethers and one ester (the ethers included di-n-octyl ether, diphenyl ether, butyl phenyl ether, phenyl cyclohexyl ether; the ester was benzyl benzoate ) was with benzyl phenyl ether. tells us then that in the This asphaltol structure there key linkages (probably benzylic are ether linkages) which are susceptible to cleavage by DTP under mild conditions. These linkages are most likely between large aggregates (or groups of molecules ) as evidenced by the significant amount of original C lost (16.2% and 23% for the 11.5 hour and
115 hour products, respectively ).
The recommendations I would make for further work in this area are as follows :
a lower stoichiometric ratio of DTP/asphaltol (the ratio (1 ) Use I used was about 23/1 ) to see if any changes are apparent from the results reported here,
(2) Develop a chromatographic separation technique so that the material cleaved from the asphaltol may be separated from the . triphenylphosphine oxide for analysis and study, and
( 3 ) During work-up of the /as DTP phaltol reaction mixture,
I would add the water (maybe even increase the amount of water added) and allow the mixture to stir for 24 hours to see if the hydrolysis can be accomplished in the same reaction vessel. 92
References
1. r is L. and Grandy, D.W; J . m. 57 Pet ak , Che &i. �, (689 ). 2. wry, 'Chemistry Supplemental Lo H:- H. &i.; of Coal Utilization•, Volume, Wiley & Sons , New �ork, 1963 , p. 262 .
Ig iak, and Gawlak, Fuel 3. nas B .S. M.; �, 56 ( 21 6 ).
4. Ignasiak, B.S. , Fryer, J-. F ., Jadernik P.; Fuel 1978, 57 (585 ) . , - -- -
5. I asiak, B.s., et Fuel (578). gn __al·, �, 57 6. Ignasiak, Fuel 1980, 59 (757 ) B.s., __, . et al· - -- -
7. achowska, H.; Fuel 1 979, 58 (99 ), A : W - - - C -91 : 1 25757n .
8. ch , Hirosawa, r t , M.; Ou i K., K., Mo i a Nenryo Kyokaishi :41 73 7t . �, 21. (765), CA : 2.!_ Winans, Haysatu, Sc t, R.G., Moore, 9. R.E., R., ot L.P. Studier, p , M.H.; Pre arative Coal Chemi stry Wo rkshop 1976, (174-88 ) &iited by Peters, H.M. and Ross, D.S. CA: 90: 124259u . 10. Yoshii, T., Y.; Kyokaishi 58 797-800). Satou, Nenryo (629; CA : 2,S:200705v 1..21?,
1 1 . Mayo, F .R., Kirshen, Fuel (405 ). N .A.; 121.§, 21. 12. Wi ser, W.H.; 1 Scientific Problems of Coal Utilization ', #L6 DOE Symposium Series, Cooper, B.R. &iit., 1978, p. 21 9.
1 3 . Chakrabartty, S.K. and ret chm r, H.O.; (160-3 ). K s e � � 21_ 14. Burwell, Chem. Rev. 1954 , 54 (61 5). R.L.; - - ,,..-_ -
itw t , Stre ei ser, A. Jr ., Hea hcock C.H.; •Introduction to Organic 15. 0 Chemistry' , MacMillan, New York , 1976, p. 239-4 .
Jung, M.E. and Lyster., Or�. (3761 ) . 1 6. M.A.; .:!..• �· !2Z?'l!£ 1 7. Mayo, F.R., Buchanan, D.H., Pavelka, L.A.; 1ACS Divi sion of Fuel h mistry 182 (1980). C e Vol. 25' , #2, pp D. ; 18. Hoffman, H., Horner, L., Wi el, H.G., Michael, ). �· �· !.2.£?, ..25. (523 19. Schaefer, and Higgins, (1607 ). J . P . J.; .:!..• Org . Chem . 1.@,__l?_
Burton, D.J. Koppes, Chem. {3026 ). 20. and W.M. ; .:!..• Org . �, Jill_ 93
Anderson, A.G. Jr. Freenor, F.J.;�!!_. Chem ( 626). and Org . . �, ;J1 r d C.J. 22. 'Aldrich Library of Inf are Spectra•, 2nd Edition, Pouchert Edit.;i Aldrich Chemical Co., Milwa1lkee, 1 975 . 23. Kosolapoff, G.M and Maier, L.; ' Organic Phosphorus Compounds •, Vol. III, Wiley-Interscience, New York, p. 369 .
24. 1 Aldrich Library of NMR Spectra•, 1st Edition, C.J. Pouchert Edit., Aldrich Chemical Co., Milwaukee, 1 974 .
25. Karr, c. Jr .; 'Analytical Methods For Coal and Coal Products•, Academic Press, New York, Vol. I p. 203, 1978 .
26. Denny, D.B., Denny, D. z . , hang, B.R.; if.• !!!• Chem. Soc . (6332 C J..22§, .2Q. ). 2 C.H.; 1 . Streitweiser, A. Jr . and Heathcock, •Introduction to Organic Chemistry•, MacMillan, New York, 1 976, p. 906 .
28. Karr, C.J.; 'Analytical Methods For Coal And Coal Products•, Academic Press, New York, Vol. II p. 76, 1978.
29. Brown, J.K., Ladner, W.R., ( 7 , 87 ) Sheppard, N.; Fuel 19-60, 9 . - .32.. 30. Starting with 0.5026 g of asphaltol 252-B (MW = 1 040 ) during the extraction a total of g of III.A -a was recovered. of o.422mat5 erial (i} was calculated The apparent MW the lost (0.801 g) to be 770. n 94
Vita -
Name : Michael L. Ballard
B Place of irth : Vincennes, Indiana
Date of Birth : January 1 5, 1 954
Secondary Education and Degrees
F.a.stern Illinois University
Charleston, Illinois
1 972-1 976 B.s. Chemistry
Aldrich Chemi cal Company
Milwaukee, Wisconsin
1976-1 979 Synthetic Organic Chemist (Production )
F.astern Illinois University
Charleston, Illinois
1979-1 981 , M.S. Chemi stry