the role of aliphatic and aromatic coal structures and macerals in low

317 Pages·2006·4.59 MB·English

The Role of Aliphatic and Axomtic coal. Structures and Macerals
in hxi Tempratwe oxidation processes
Yonqsarq Yun, mum ~akab,W illiam H. Mcclennen, George R. Hill
and Henk L.C. Meuzelaar
Bimaterials Profiling Center, University of Utah
Salt Lake City, Utah 84108
It XRXUCTION
LKW temperature (5 ~OOOC)c oal oxidation processes are kncwn to have a pro-
nomced effect on important mal properties such as coking and caking behavior
(1,2,3), heat of ccrmbustion (3), flotability (2) slurry pH (Z), tar yield and ex-
tractability (3). However, the mcharu'm i and kinetics of the chemical reactions
underlying such "eathering" processes in coals are far f mw ell understood. Among
the most pdsing analytical methcds for elucidating the structural and cam-
positional changes during coal weaulering are Fourier Transform i n f d ( FTiR)
spectmscopy (4!5) and pyrolysis IMSS spectm+q (Py-?) (6,7). FTIR studies
performed by Pamter et al. (8) shaw a strong mcrease 111 cazt3onyl as well as ester
moieties. Recent ~~-WZj&mntsb y JW et al. (9) have danonstrated a pro-
nzoLun.&- .-2dedcr>e a.s eL -i,n=- t=h*e ppyrr-o.Glybs&iisG yr,i eozld ;s o-f p&hpeniao.l ic. (a$zndw nxaipyhyitihGaalen- dn ic& w@npinCen)t s
short chain aliphatic moieties. Fy-MS techniques based on the ocrmbination of Curie-
paint pyrolysis, low voltage mass spctmmeternym anpd conputerized multivariate data
analysis methcds were found to be sensitive to detect changes in bituminous
and lower rank coals after exposure to air at 80 C for d y 1 or 2 days (9). More-
over, time-resolved Fy-E techniques ca le of recording up to 10 spedra per
second at heating rates of appmx. 100%~ were shown to pmvide valuable inform-
ation about the med.lanisms and kinetics of vacuum pyrolysis processes in fresh and
weathered coal (9). Meuzelaar et al. (3) interpreted the pronounced decrease in
extradable al&lnaphthalene mieties in hi% volatile B bituminous Hiawatha mal as
a loss of 9nobile phasetqw npnents by attaclnnent to the "network phasev*. under the
conditions employed in those experiments, such f'graftingf' processes appear to be
more prevalent than the often hypothesized crosslinking reactions between ma-
molecular camponents. These interpretations find support in recent studies by
Larsen et al. (lo) on the influence of weathering processes on coal swelling prop-
erties in solvents.
In the present article new Py-W data on low tapratwe oxidation effects in
several U.S. coals of different rank and origin will be discussed with special
-is on the role of different coal mcerals.
m m m
'~hrees eparate experiments were performed: (a) oxidation under dynamic reactor
flow conditions; (b) oxidation under static conditions; and (c) pyridine extraction
of fresh and oxidized coal.
Mer dynamic conditions, 20-24 g aliquots (-60 mesh) of four 4 s (W yddlc,
Maville $6, Hiawatha, Upper Fmeprt) Were exposed to an air flaw of 5.10 lnl/ini.n in
a 100 ml glass reactor, at 100°C for 170-212 hours (11). Under static conditions,
f- m-1 concentrates were Oxidized with air, viz. vitrinita and d-fusirdte
fractions prepared by density gradient centrifugation of Dakota seam coal (New
Mexico, psoc 858), sporinite prepared by density gradient centrifugation, f m
Brazil block seam, Indiana Fastern (FSCC 107) and a hand-picked resinit.e sample f m
the wattis seam (Utah). five ng aliquots of these mcerdls were to
air at 100°C for 100 hours in sealed 2 1 flasks.
pvridFne extraction was performea on Hiawatha coal in a m a r m e t e x-
tractor for 24 hours at rooon temperature. A weathered coal sample for midine
129
extraction was prepared by exposing a 15 g aliquot (-60 mesh) to air at 100°C for
212 h- (12) under dynamic flow ocoditions.
~ lwih ole codl samples as well as insoluble extraction residue were hand ground
der nitrogen a m e r e i nto a fine, uniform suspension in spectrpsraae methanol
(5 mg of -le per ml of methanol). ’Ibe pyri- cted fractlon was dissolved
inamucture of benzene and methanol (3:1, v/v) at a mnoentratim of 2 mg/ml. Five
microliter drops of the above suspensions and the solution were coated on.the ferm-
magnetic filaments used in curie-point pyrolysis and ajr-dried un3.e.r amtmwus
rotation, dtirq in approximately 20 q of dry sample on the filamerrt.
pyrolysis MS runs were performed under the same d t i o n s f or all samples,
i.e., electron ~ergy12 eV, mass range 20-260 m/z, scannbq0rate 1000 amu,/s,
temperature rise time about 6 s, equilibrium V t w e 6 10 C, total heating time
10 s (13,14). In Order t0 WalUa- Weathering-XdUCd differences, ccanputerized
dtivariate analysis techniques such as factor analysis and discriminant analysis
were applied to the Ms data (14,15).
T(EsuLTs AND DISCUSSION
LDJ voltage pyrolysis mass spectra of fresh and weathered coals are Shawn in
Figures 1-4. Ccqyison of the spectra of the four fresh coal samples in FirJ””
la-4a the obvious effects of rank differences between the two subbhmunous
coals (Adaville #6 and .y lodak) and the two high volatile bi- coals (Uppr
Freeport and Hiamtha) Whereas the spectra of the former (figs. la and 2a) are
daninated by dihydmxybenzenes and phenols of various degrees of alkylsubstitutim,
the spectra of the latter (figs 3a and 4a) show a dramtic decrease in d@dmxy-
benzene signals coupled with a markedly higher abudame of ammatic hy&xc&xm
series such as naphthalenes and benzenes. mis trend appears to be most proolcunced
in the Upper Freeport sample, perhaps due to its slightly higher rank (bozderline
hvAb/mvb) than the Hiawatha coal (hvEb/hvAb). Howw-e r, Ure,ma-jor differences in age
and origin between t-h e two samples (vpper Freeport carbom femus, Interior
province; Hiamtha Qretaceous, Rocky Mountain province) should also be kept in
mind. As noted by Given et al. (16) Western coals tend to have 1- hydmxyl
content than Interior dKtern province coals of ocmrparable rank. For a detailed
discussion of rank effects on the law v-oltag e pyrolysis mass spectra of 102 Western
U.S. coals the interested reader is referred to previous pb+iations (6,7).
The differences between the spectra of the two subbitmumus coals (figs. la
and 2a) are far less pmnounced. Again a snall rank effect. my be present. On the
basis of equilibrium moisture content wyodak coal sanples could classified
as lignites (17) whereas the rank of the Adaville #6 coal is subbitumvmus B.
Although the pyrolysis mass spectra of western lignites and SuhbitLmLincrus coals do
not shm markd differences (7) , the slightly higher relative abundance of the
naphthalene series in the Adaville coal (Figure 2a) is rather typical as is the
sanewhat different distribution of phenols.
It should be pointed out here that chemical identification of the different
peak series in the low voltage pyrolysis mass spectra of coals is tentative only and
is based laxyely on supporting data obtained by GC/MS (18,19), m/MS (20) and NMR
(21) studies of similar coal pymlyzates. ~ ~ w esirnce, e very mass peak is likely
to contain mntrihtians frm more than one chemical ccanpomd, the use of chemical
labels is only meant to identify those conpnents thought to contribute mcst
strongly to the respective peak series.
The effects of lm tenperatwe oxidation on the four coals can be examined by
llsubtractingll the spectra of the weathered coal ~ l eins Fi- Ur4b frcw those
of the correqom3irg fresh samples using a *murant analysis pruodure (9)
(Figures la-4a). This results in the discrlrmnant spectra shown in Figures lc4c.
ccrmparisun of the latter reveals an intriguing, generalized trend. ‘Ibe -tic
difference in susceptibility to weathering effects between those structural moieties
130
prOaUcing the abarementioned aliphatic peak series and those mieties giving rise to
the annnatic hydrocarbon and hydmxyaromtic peak series.
A second, intriguing observation fnnn Figures lc-4c is the qualitative sim-
tlarity between the low m o l d a rw ei@t ion signals found to increase upon weather-
W. In all four coals these peak series appear to rep-t highly polar short
chain aliphatic ccanpounds such as a+*(m /z 281, a2+(*m/z 441, and CH COOH+* (m/z
60). Furthermore, it is interesting to note the high intensity of m/z 32, thought
to represent increased retention of the methanol solvent (wed to prepare mal
suspensions for Fy-hE analysis) in the more polar, weathered coals.
First, let us colloentrate on the differential loss of aram3tic % aliphatic
pyrolyzate cconponents. It is ten@- to try to explain this on the basis of
possible differences in the susceptibility of the different coal m a d m npnents.
As evident fnnn the averaged pyrolysis mass spectra of mceral concentrates in
Figure 5, marked differences exist between the relative aliphaticity and annnaticity
of pyrolyzates from fusinites (highly ammatic), vitrinites and sporinites
(prkily aliphatic). Other liptinitic mads such as alginite and cuthite
provide even stronger aliphatic Fy-MS patterns than sporinite (22), whereas
resinites fmm Western U.S. coals are characterized by alicyclic and hydrmmnatic
series (23). Thus, in principle it would be possible that lw temperature oxidation
could effect primarily the more highly aromatic vitrinitic w o r f usinitic mcerals
while leaving the more aliphatic mcerals relatively unaffected.
This brings up the interesting question which of the various liptinites are
primarily responsible for the renarkably similar aliphatic patterns observed in all
four coals in Figures 1-4. Althougfi sporinite is likely to contribute to the Fy-MS
pattern of the carboniferous Upper mrt coal, cretaceaus (Adaville, Hiawatha)
and Tertiary (Wycdak) Western coals generally contain only minor amounts of
sporinite. Many Western coals, h-er, are knmn to contain marked ammnts (e.g.,
5-10%) of resinite (23). Nevertheless, typical Fy-IS patterns of resinites are
quite different from the prcnninent aliphatic h v npat terns in Figures 1-4.
Consequently, the authors believe that two ubiquitous liptinitic macerals, namely
alginite (or detrital forms such as b i M t e ) and cutinite are primarily respon-
sible for the observed aliphatic Py-m patterns and may well be present in much
larger quantities than estimated by m.iQppetrographic techniques. Evidence
supporting the potentially important contributions of algal coal ocnnponents has been
presented (thou@ not capletely interpreted) in previous ~y-ms tudies of over 130
U.S. d s (7 ,17), whereas the grossly underestimated mle of cutinitic mcerals has
recently been highlighted in several studies by De Leeuw and mrkers (24).
Fyridine extraction of Hiawatha coal before and after low temperature oxidation
(Figure 6) not only shcm a dramtic decrease in extraction yield (22%. 4%) but also
reveals a nearly ccnnplete separation between arcmatic peak series (donurant in the
extract) ad aliphatic peak series (retained in the residue). Althou@ it should be
pointed aut that the residue is likely to contain mre or less himy cca7densed
aromatic ocmponents as well (which m y well have remained I1invisible1t' o the Py-m
procedure), the apparent insolubility of the aliphatic hykccarbn conpnents is in
line with the previous contention that alginite and/or cutinite derived macerals are
likely to be responsible for the bulk of the obsenreed aliphatic n -h s+gnals
in ~y-mp atterns of whole d.Bo th types of macerdls are notoriously difficult
to dissolve (cutinites and algal kercgens can be purified by means of strong
acids!).
careful &tion of Figure 6 also provides an impOrtant clue with regard to
the possible fate of the disappearing ammatic moieties during law temperature
weathering. Fy-MS analysis of the pyridine residue of the weatherd coal sample
reveals a clear series of allylnapthalenes, apparently liberated by the vacuum
mlysis wnditions. It is tarpsting to speculate that these arcmatic moieties
became 1- (l'grafted") to the mcrmnoldar network during the weathering
p m . ~o~aretrh,e possibility of physical entramt (clathration) cannot be
out either. present indications are that the grafted (or trappea) arcnnatic
moldesal one cannot fully explain the markd loss of aromatic and hci-t
moieties but that s~mear crmatic rbgs are destruyed by the oxidation process (in
spite of the relatively law temperatures used), resulting in the formation of the
131
short chain aliphatic acids and ketones seen in Figures lc-4c. Further for
this exp1anation is fcrund in the remilts of law tanperature oxidation s u e s o n
selected m a d c cmcentrah of similar rank shayn in Figures 7 and 8.
!Ihe factor score plot in Figure 7 meals that the resinite and semifusinite
mcenmtes underwent more swere oxidative changes, which resulted in the forma-
tion of oxygen contajning aliphatic moieties, than the vitrinite mamtxate. The
sporinite concentrate showed the least severe weathezing-- degrada?~. m s
sugyests that the abundant hydroammtic and aromatic moieties rn the resmte and
semifusinite samples respectively, prnvi.de the most likely source for the short
chain, carboxyl- and carkunyl-containing, aliphatic caqmn~%ob served in the FY-MS
patterns of weathered mdl.
conclusion, aliphatic, hydmxabn-rich coal st~~ctur t-es to be de-
rived. fmn cutinitic or algal sources show little or no chemical changes under law
temperature oxidation corditions whereas vitrinite and sprinite &aw a mo"dOeraMte ;,
oxidation tendency. oxidation experiments with d san d m a d
trates f m d ifferent saurces will have to be carrid out in order to deterrmne
general validity of the obsenred phemaneM for coals of different rank and aeposi-
tional history.
Am-
The authors gratefully admawledge gifts of fresh upper Freeport and wyodak
samples by Dr. Karl Vorres (Argcome National Laboratory) ; of fresh Admille #6 ard
Kiawatha samples by Utah Payer and Uqht and Of rnacerdl ConCentratesby Drs. R.
pwSmire and J. Karas (university of Utah). The work reported here was supported by
D3E COn,trct DE-FG22-84K70798.
REFERENCES
w,
1. crellbg, J.C., R.H. schrader, and L.G. Benedict, 58, 542 (1979).
2. Gray, R.J., A.H. Rhoadesi and D.T. King, AlME Transactions, 260, 334-341 (1976).
3. Mewelaar, H.L.C., W.H. McClennen, C.C. C&y, G.S. Metcalf, W. Windig, J.R.
T h w , a nd G.R. Hill, in ACS Preprints (Div. of Fuel Chem.), 29 (5), 166-177
(1984).
4. R h h , C.A., J.T. Senftle, M.M. Davis, and P.C. Pam, E,62, 1387 (1983).
5. cro~uer,D .c., R.G. -, R.S. ~.=nkin~A,. Davis, and P.C. painter I Fuel,
62, 1124 (1983). w,
6- Harper, A.M., H.L.C. MeUzelW, 63, 793 (1984).
7. MeuzdaW, H.L.C., A.M. Hazper, G.R. Hill, and P.H. Given, Fuel, 63, 640-652
(1984).
8. Pain*, P.C., R.W. Snyder, D.E. Pearson, andJ. Riong, Fuel, 59, 282 (1980).
9. Jakab, E., E. Hoesterey, W. Windig, G.R. Hill, and H.L.C. Meuzelaar, "Effects of
Law Temperature Air aidation (Weathering") Reactions on the srrolysis Mass
Of U.S. Coals~s~u bmittd to Fuel (1986).
10. larsen, J.W., D. Lee, and T.E. SchinipThe Effect Of Atwspheric Oxygen on the
Of codls", Final Rep& to GKt, CCntract NO. 5082-260-0907 (1985).
11. Meuzdam, H.L.C., G.R. Hill, Y. Yun, and W. Windig, "Weathering Effects on
S ~ I aUnd R~eac tivity of US. coalsf1, Third Quarterly Report to DOE, Contract
NO. m-FG22-84K70798 (1986).
12. fill, G.R., H.L.C. Meuzelaar, J.M. Futrell, A.M. Harper, and. W.H. McClennen,
"Role of preasphaltenes in coal conversion ReactionsqgF, inal Report to DOE,
ccoltract NO. DE-FG22-82pc50970 (1984).
13. MeUZdaar, H.L.C., J.H. Haverkanp, and F:D. H i l m , qrl.olysiS of
Recent and Fosssil Bicnnaterials: ccnnpenllu m and ,. -saAlt
Arrsterdam (1982).
14. MeUZdaar, H.L.C., W. Windig, A.M. Harper, S.M. Huff, W.H. MCC2-, and J.M.
Ri-, Science, 226 (4672), 268-274 (1984).
15. Windig, W. and H.L.C. Mewelaar, Anal. Ulem., 56, 2297-2303 (1984!.
16. Yarzab, R.F., A.. AMel-Baset and P.H. Given, Gecchnu' ca et l3Xmdun'c a Acta, 43,
281-287 (1979)
132
I
17. Metcalf, G.S. W. Windig, G.R. Hill, H.L.C. Meuzelaar, Characterization of U.S.
Lignities by pyrol ysis Mass spectrcanetry and Eailtivariate Analysis, coal
Geolcgy, (1986) in press.
18. Blazso, M. and E. Jakab, J. Anal. Awl. Fyrol., 8, 189 (1982).
19. Nip, M., J.W. de Keeuw, P.A. Schenck, H.L.C. Meuzelaar, S.A. Stat, P.H. Given,
and J.A. Boon, J. Anal. Appl. Ml., 8, 221 (1985).
20. Meuzelaar, H.L.C., W.H. Mcclennen, J.H. TcniLinson, and D.L. Pope, Proc. Int.
Cnnf. on coal science, ~xlsseldorf,p p. 816-821 (1981).
21. Hceshy, B.L., W. W m g , H.L.C. Meuzelaar, E.M. Eyring, D.M. Grant, andR.J.
mgmire, "An Integrated V i cAppr oach to the chemical characterization
of Fyrulysis Oils" ACS Freprmts (Div. of Fuel QMn.) Denver 1987.
22. larter, S.R. and A.G. Douglas, J. -1. -1. ml., 4, 1-19 (1982).
23. Csellhg-, J.C., personal conmumication.
24. Nip, M., E.W. Tegelaar, H. Brinkhuis, J.W. de Ieeuw, and P.A. Schemk, Analysis
-of Recent and Fossil Plant Cuticles by Curie-point qrl-olysis-Gas Olrmnatography
and Curie point pYr0lysis-qa.s chmmatography- Mass spectrcmretry: R€ccqn'l tim of a
New, HimyA liphatic and Resistant Biopolymer, in Adv. in Org. Geochan., (1986)
m press.
25. Meuzelaar, H.L.C., Harper, A.M., *re, R.J., and Icuas, J., Int. J. coal
Geolq, -4, 143-171 (1984).
!
133
134
135
mL w
c , L
\N
1 -
O h 0 - A i l - .
3 5
N
cU, E
cU,
a
VI a
U VI
01 U
mm
mm
0) w
m
I m
v
0)
.r
E
.r
L
=.c-. ,
n
- -.
n N
A%!SUJ?UI UOI Lelol %
136
c
J:
0,
.r
rcLW,
m
W
Y
0 m
.ELr - .LE- - EO P
a c
-00
vw.umr vVwI+LoJ
VmI- o
W L
u3 LurOn
-w,0...o .Er VN
cVI,- uVEI
E
GEI
E
n a
E
0 V0
m
, j
-.w0cw.)'cw*)ow-a.
L L L L
w w w w
r c c c
+mJ+mJCm,+mJ
%!%!%!%!
V V Y V
P O 0 0
CHARACTERIZATION OF SURFACE FUNCTIONALITY OF COALS BY
PHOTOACOUSTIC FTlR (PAIFT) SPECTROSCOPY, REFLECTANCE INFRARED
MICROSPECTROMETRY, AND X-RAY PHOTOELECTRON SPECTROSCOPY (XPS)
Brian M. byllEll, Lal-lm Lancaster,
and J. Anthony MacPhee'
Department of Chemlstty, St. Francis Xavier University,
Antlgonlsh, Nova Scotla B2G 1C0, Canada
This paper illustrates detection by the technique of PhotoAcoustic Infrared Fourier Transform
(PAIFT) spectroscopy of new carbonyl-type functionality at the surfaces of powdered bituminous
coals, generated by both base-promoted and by thermal decomposition of precursor peroxide
species, postulated as ubiquitous constituents at the surfaces of all except the most freshly
prepared samples. In artificially oxidized coals, there are quantitative associations between the
level of carbonyl content revealed by PAIFT spectra and plastic properties of the coals. In
additional, related studies, we have derivatized and quantified the hydroxyl group content of the
oxidized coals by trifluoroacetylation, followed by hydrolysis and anion-chromatographic analyses
for trifluoroacetate. PAIFT and Fluorine-XPS spectra of the trifluoroacetylated species are
discussed. Maceral components and mineral inclusions have been identified and characterized in
sectioned, polished surfaces of Canadian bituminous coals using reflection FTlR
microspectrometry; this direct examination shows promise for real-time monitoring of various
reactions at surfaces.
INTRODUCTION
Sensitivity enhancements in spectroscopy show promise of specific characterization
of functional groups in coals and their chemical transformations. We have applied one of the newer
techniques (PhotoAcoustic Infrared Fourier Transform spectroscopy (PAIFT spectroscopy) ) to
generate a data base of coal samples amounting to several thousand spectra. Various preliminary
reports of our results have been published or presented (1-7).
Among the manifold advantages possessed by PAIFT spectroscopy for the study of
coals are minimal sample preparation, insensitivity of signal to degree of subdivision, and the ability
to observe Wace featu res generated by chemical transformations. PAlFT spectra of coals do not
exhibit the sloping baselines characteristic of alkali halide pellet spectra (contrast a recent report of
such FTlR spectra of Canadian coals (e))a, nd subtraction to yield difference spectra revealing
introduced features is straightfolward and avoids arbitrary scaling procedures. The first paper of this
series (7) reported detection by PAIFT spectroscopy of new carbonyl oxygen functions at the
surfaces of powdered bituminous coals, derived from base-promoted or thermal decompositions of
precursor peroxide species, which we suggested were present ubiquitously as surface entities for
all except the most freshly prepared samples. In more recent work, we have discovered that thermal
decompositions accompany a standard vacuum drying procedure.
~ ~~
Energy Research Laboratories. CANMET. Energy, Mines and Resources Canada, Ottawa. Canada K1A OG
138