Formate species in the low-temperature oxidation of dimethyl ether

Article abstract of DOI:10.1016/S0045-6535(00)00231-9

The oxidation of dimethyl ether (DME, 340 ppm in 10% O₂) has been studied experimentally in an atmospheric pressure laminar flow reactor in the temperature range from 240°C to 700°C for residence times in the range 2-4 s. The influence of nitric oxide additions up to 620 ppm to the feed gases has also been investigated. Products of reaction were determined by FTIR. In the absence of NO, reaction is first detected at about 260°C. The products in the low-temperature region include formaldehyde (HCHO), and formic acid (HCOOH). The addition of NO leads to the appearance of methyl formate (CH₃OCHO). While the overall behaviour of the system can be explained qualitatively in terms of typical low-temperature hydrocarbon ignition, recently published chemical kinetic models for DME ignition do not allow for the formation of these formate species. We find no experimental evidence for the formation of hydroperoxymethyl formate (HPMF, HOOCH₂OCHO) which is predicted by the models to be a significant stable intermediate at temperatures below 350°C. Since both formic acid and methyl formate have potentially harmful health effects, these observations may have significant implications for use of DME as a diesel fuel.

Full text of DOI:10.1016/S0045-6535(00)00231-9

Chemosphere 42 (2001) 583±589
Formate species in the low-temperature oxidation of dimethyl
ether
a
Irene Liu , Noel W. Cant , John H. Bromly , Fred J. Barnes ,
a
b
b
c
Peter F. Nelson , Brian S. Haynes
d,
*
a
School of Chemistry, Macquarie University, Australia
b
AlintaGas, Perth, Australia
CSIRO Division of Coal & Energy Technology, Australia
Department of Chemical Engineering, University of Sydney, Sydney, NSW 2006, Australia
c
d
Abstract
The oxidation of dimethyl ether (DME, 340 ppm in 10% O2) has been studied experimentally in an atmospheric
pressure laminar ¯ow reactor in the temperature range from 240°C to 700°C for residence times in the range 2±4 s. The
in¯uence of nitric oxide additions up to 620 ppm to the feed gases has also been investigated. Products of reaction were
determined by FTIR.
In the absence of NO, reaction is ®rst detected at about 260°C. The products in the low-temperature region include
formaldehyde (HCHO), and formic acid (HCOOH). The addition of NO leads to the appearance of methyl formate
(CH3OCHO).
While the overall behaviour of the system can be explained qualitatively in terms of typical low-temperature
hydrocarbon ignition, recently published chemical kinetic models for DME ignition do not allow for the formation of
these formate species. We ®nd no experimental evidence for the formation of hydroperoxymethyl formate (HPMF,
HOOCH2OCHO) which is predicted by the models to be a signi®cant stable intermediate at temperatures below 350°C.
Since both formic acid and methyl formate have potentially harmful health e€ects, these observations may have
signi®cant implications for use of DME as a diesel fuel. Ó 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Formic acid; Ignition; Nitrogen oxides; Diesel fuel
1. Introduction
ether (DME) has been proposed as a diesel fuel; it has
been shown that DME has most of the desirable prop-
erties for a substitute diesel fuel: a cetane number (>55)
close to diesel; comparable eciency; and, most impor-
tantly, greatly reduced smoke emissions and signi®cantly
reduced NOx. The overall emission levels from DME
fuelled diesel engine tests have been shown to surpass
the Californian ultra low emissions vehicle (ULEV)
regulation for medium-duty vehicles (Fleisch et al.,
1995).
Oxygenated fuels such as methanol, methyl ethyl
ketone and methyl isobutyl ketone have been investi-
gated as alternative transportation fuels. The impetus
for these investigations has been largely based on the
generally lower pollutant emissions obtained with these
fuels (Bowman and Seinfeld, 1995). Recently dimethyl
* Corresponding author. Tel.: 61-2-9351-2470; fax: 61-2-
The possible future use of DME as a fuel has resulted
in a number of studies of its potential atmospheric re-
actions, and of its oxidation (Japar et al., 1990; Jenkin
et al., 1993; Sehested et al., 1996, 1997; Maricq et al.,
9351-8266.
E-mail address: haynes@chem.eng.usyd.edu.au (B.S.
Haynes).
0045-6535/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 4 5 - 6 5 3 5 ( 0 0 ) 0 0 2 3 1 - 9

584
I. Liu et al. / Chemosphere 42 (2001) 583±589
1997). Under conditions relevant to atmospheric pro-
cesses, the reaction of O2 with the methoxymethyl rad-
ical is the key to the formation of products. This
reaction proceeds via the chemically activated adduct to
form either the stabilised peroxy radical CH3OCH2OO
or, at lower pressures and in the absence of collisional
stabilisation, to formaldehyde (Sehested et al., 1996,
1997; Maricq et al., 1997):
temperature ignition of hydrocarbons, proceeding via
peroxy intermediates whose further reactions are capa-
ble of leading to chain branching:
R´ ‡ O2 $ ROO´ ! ´QOOH ! Á Á Á ‡ 2OH ꢀtypically†
RH ‡ OH ! R´ ‡ H2O
Dagaut et al. (1998) observed a characteristic region of
negative temperature coecient (NTC) consistent with
this overall mechanism when the increasingly unfa-
vourable equilibrium in the reversible formation of the
peroxy species leads to inhibition of the overall reaction.
Reactions between NO and (organic) peroxy radicals
have been found, recently, to be responsible for the
mutual sensitisation of the oxidation of the parent or-
ganic and of the NO (to NO2) under low-temperature
thermal oxidation conditions (Bromly et al., 1992, 1996;
Hjuler et al., 1995; Doughty et al., 1996). Alzueta et al.
(1999) concluded that the reaction of peroxy species with
NO at low temperatures may be important but did not
identify product details in this temperature range.
In this work, we report the results of an experimental
study of the low-temperature oxidation of DME in a
¯ow reactor. The reactor was operated at atmospheric
pressure, and at temperatures of 240±700°C. Oxidation
of DME (340 ppm) in a large excess of oxygen (10 mol%
O2) was studied in the presence and absence of NO.
While these conditions are not representative of actual
combustion conditions, they allow key processes in the
overall reaction to be identi®ed under essentially con-
stant oxidant concentration. The results suggest that the
low-temperature oxidation, and, by inference, the igni-
tion, of DME may occur through a more complicated
process than previously believed.
This description of the kinetics of the reaction
O2 ‡ CH3OCH2 has been found to apply at tempratures
up to 473 K and pressures up to 18 bar (Sehested et al.,
1997).
Under atmospheric conditions, the self-reaction of
CH3OCH2OO radicals leads to the formation of methyl
formate, CH3OCHO, while reaction with HO2 produces
CH3OCH2OOH (Jenkin et al., 1993). Methyl formate is
also formed in the presence of NO (Japar et al., 1990):
CH3OCH2OO ‡ NO ! CH3OCH2O ‡ NO2
CH3OCH2O ‡ O2 $ CH3OCHO ‡ HO2
CH3OCH2Oꢀ‡M† $ CH3OCHO ‡ Hꢀ‡M†
ꢀ2†
ꢀ3†
ꢀ4†
Here it should also be noted that formic acid, HCOOH,
has also been found in room temperature reaction
products of CH3OCH2OO (Japar et al., 1990).
DME oxidation has also been studied in a jet stirred
reactor from 1 to 10 atm and 550±1200 K (Dagaut et al.,
1996, 1998). In contrast to the major products reported
at room temperature under atmospheric conditions,
methyl formate was not reported. A comparison be-
tween the observed products and a detailed kinetic
model suggested that DME reacts mostly via the ther-
mal decomposition of CH3OCH2
2. Experimental
DME was oxidised in the presence and absence of
NO from 240°C to 700°C. The gases were prepared by
diluting 1% DME/He, 3600 ppm NO/He, and air (BOC
zero grade) in He to the required concentration using
Brooks mass ¯ow controllers. The mass ¯ow controllers
were calibrated with a Buck MP5 soap ®lm ¯ow meter.
The combined ¯ow was 200 cm3/min STP.
CH3OCH2ꢀ‡M† ! CH3 ‡ HCHOꢀ‡M†
ꢀ5†
Alzueta et al. (1999) also found reaction (5) to be im-
portant in their ¯ow-reactor study of DME oxidation at
temperatures from 600 to 1500 K. However, they did
®nd evidence for alternative low-temperature ignition
pathways in the lower range of the temperatures they
studied.
The reactor consisted of a 6 mm inner diameter
quartz tube which was heated by a three zone laboratory
tube furnace (Lindberg, Model 55346). The bulk of the
gas was preheated by passing through the length of
the reactor in a separate quartz section; the DME was
added through a second inlet to prevent premature de-
composition. The residence time for the gases at 240°C
was 3.9 s.
Recently, e€orts to construct detailed chemical
models of low-temperature DME ignition and oxidation
have been published (Curran et al., 1998; Dagaut et al.,
1998; Alzueta et al., 1999). These models are built
around processes typically occurring in the low-
The gas mixtures leaving the reactor were fed to a
Digilab FTS-15/80 FTIR spectrometer for analysis in a

I. Liu et al. / Chemosphere 42 (2001) 583±589
585
multipath infrared cell of 3.2 m pathlength. The spec-
1
trometer was operated at a resolution of 0.25 cm and
the spectra were obtained by co-adding 64 inter-
ferograms. The reactants, DME and NO, and products,
CO, CO2, HCOOH, HCHO, CH3OCHO, CH4 and
NO2, were identi®ed from their infrared spectra. DME,
HCOOH, and CH3OCHO were determined from the
intensities of individual lines calibrated against reference
spectra generated from a commercial database of high-
resolution gas phase spectra (Infrared Analysis). Con-
centrations of CO, CO2, HCHO, CH4, NO and NO2
were determined by ®tting against spectra generated
using the HITRAN database and the MALT program
(Grith, 1996).
Closure of the carbon balances determined from the
experiment depended on the product composition. At
high temperatures, where the major products were CO
and CO2, balances were within about ‹10% of those
based on the inlet DME concentrations. At lower tem-
peratures, the carbon balances were poorer and the total
carbon in the products exceeded the nominal inlet car-
bon by up to 20±30%. These uncertainties are believed to
arise from the inaccuracy of the inlet concentration of
DME (based on the supplierÕs certi®cation), and the
concentrations determined by FTIR for HCHO,
HCOOH and CH3OCHO. FTIR measurements for
these species are a€ected by overlapping features and by
the quantitative accuracy of the concentrations assigned
to the reference spectra. The improved balances ob-
served at higher temperatures are believed to be due to
the more accurate method used to calculate CO and
CO2.
Fig. 1. Concentration of DME (initially 340 ppm) remaining
after oxidation in 10% O2, as a function of temperature. Resi-
dence time 3.9 s at 240°C. r ± without NO in reactant mixtures;
e ± with 350 ppm NO in reactant mixtures.
3. Results
Fig. 1 shows the extent of decomposition of 340 ppm
DME in 10% oxygen, with and without NO, as a
function of temperature. The residence time is 3.9 s at
240°C, decreasing at higher temperatures. In the absence
of NO, reaction is ®rst detected at a temperature of
260°C. The extent of reaction increases initially with
increasing temperature but reaches a maximum at a
temperature of approximately 360°C after which a re-
gion of negative temperature coecient (NTC) is en-
countered. Between approximately 500°C and 600°C,
there is very little loss of DME, but above 650°C, the
DME is almost entirely removed.
With the addition of 350 ppm NO to the reaction
mixture, the decomposition pro®le is greatly altered.
Removal of DME is now inhibited and begins only at
temperatures above 320°C; there is no NTC region and
DME consumption is complete at temperatures in excess
of 550°C.
Fig. 2. Major products from oxidation of 340 ppm DME in
10% O2 as a function of temperature. Residence time 3.9 s at
240°C. s ± CO; d ± CO2; j ± HCOOH; Ã ± HCHO; m ± CH4.
In the absence of NO (Fig. 2), the region of low-tem-
perature reactivity gives rise to CO, formaldehyde
(HCHO), and formic acid (HCOOH) in similar pro-
The products of reaction corresponding to the de-
composition pro®les in Fig. 1 are shown in Figs. 2 and 3.

586
I. Liu et al. / Chemosphere 42 (2001) 583±589
sion of NO to NO2. At higher temperatures, T > 550°C,
NO again appears in the products and there is negligible
NO2 above 700°C.
Fig. 4. Conversion of NO to NO2 as a function of temperature
for conditions shown in Figs. 1 and 3. , ± NO; . ± NO2.
Fig. 3. Major products from oxidation of 340 ppm DME in
10% O2 plus 350 ppm NO as a function of temperature. Resi-
dence time 3.9 s at 240°C. s ± CO; d ± CO2; j ± HCOOH; Ã ±
HCHO; n ± CH3OCHO.
portions. The peaks in the CO and HCOOH pro®les
occur at the critical temperature at which the con-
sumption of DME is at its maximum, after which the
concentrations of these species decrease to zero. Formic
acid is not seen again in the products at higher tem-
peratures but CO becomes a major product in the region
of high reactivity at T > 600°C. The concentration of
HCHO reaches a maximum at a slightly higher tem-
perature than CO and HCOOH and this species is still
present, at low concentration, in the products in the
relatively unreactive region above 500°C. Above 600°C,
some CH4 and CO2 also appear in the products.
When NO is added to the reaction mixture (Fig. 3),
the increase in consumption of DME is seen in the
products mainly as CO and, increasingly above 600°C,
as CO2. There is also a signi®cant concentration of
HCOOH and of CO2 in the low-temperature region,
below 500°C. At temperatures between 300°C and
400°C,
a new product appears: methyl formate,
CH3OCHO. The concentration of HCHO is lower than
in the absence of NO but nevertheless signi®cant, espe-
cially around 450°C. No CH4 is detected in the products
when NO is present.
Fig. 5. Products from oxidation of 340 ppm DME with 10% O2
at 320°C, as a function of NO inlet concentration. Upper panel:
r ± DME remaining, , ± NO remaining, . ± NO2 formed.
Lower panel: s ± CO; d ± CO2; j ± HCOOH; Ã ± HCHO; n ±
CH3OCHO.
The onset of reaction of DME in the presence of NO
is seen in Fig. 4 to correspond to the complete conver-

I. Liu et al. / Chemosphere 42 (2001) 583±589
587
DME. The formation of HCHO is initially promoted
but reaches a maximum at quite low values of [NO]0 and
is negligible at high [NO]0. On the other hand, HCOOH
and CO2 formation, which are also initially promoted by
NO addition, remain important at the highest levels of
[NO]0 employed.
4. Discussion
The results for the consumption of DME in the ab-
sence of NO are qualitatively similar to those reported
by Dagaut et al. (1998). In both cases, there is a clear
region of negative temperature coecient (NTC) at
temperatures above about 360°C. Below the turnover
temperature, low-temperature hydrocarbon ignition
processes may be presumed to dominate the system.
Recent attempts to produce comprehensive kinetic
models DME oxidation include the low temperature
pathways initially proposed by Curran et al. (1998)
CH3OCH2 ‡ O2 $ CH2OCH2OO
ꢀ1b†
ꢀ6†
CH3OCH2OO ! CH2OCH2OOH
CH2OCH2OOH ‡ O2 $ OOCH2OCH2OOH
OOCH2OCH2OOH ! HO2CH2OCHO ‡ OH
HO2CH2OCHO ! OCH2OCHO ‡ OH
OCH2OCHO ! CH2O ‡ HCOO
ꢀ7†
Fig. 6. Products from oxidation of 340 ppm DME with 10% O2
at 440°C, as a function of NO inlet concentration. Symbols as
for Fig. 4.
ꢀ8†
ꢀ9†
Figs. 5 and 6 show the in¯uence of the inlet con-
centration of NO on the oxidation of 340 ppm DME
and the formation of products at 320°C and 440°C, re-
spectively. At 320°C, addition of NO gradually increases
the overall consumption of DME until the ‰NOŠ0=
ꢀ10a†
This reaction sequence explains the appearance of the
NTC region (via the reversibility of reactions (1b) and
(7)). However, it does not account for the appearance of
formic acid, HCOOH, observed here (Fig. 2) as a major
product up to the turnover temperature. Formic acid
has been reported as a minor product in the low-tem-
perature (350±420°C) oxidation of formaldehyde (Hay
and Hessam, 1971) but this is an unlikely source in this
case as the concentration of formic acid is often com-
parable to or greater than the concentration of formal-
dehyde. An alternative, slightly exothermic (DH ꢁ 30
kJ mol 1) route for reaction 10 may, however, allow for
the observation of formic acid
‰DMEŠ ratio approaches unity; beyond that, further
0
additions of NO inhibit DME consumption and the
reaction is completely suppressed at ‰NOŠ0=‰DMEŠ
0
>1:5 (Fig. 5). Essentially complete conversion of NO to
NO2 occurs at ‰NOŠ0=‰DMEŠ < 1; there is no NO2 in
0
the products once DME consumption is suppressed. The
extents of formation of CO, CO2 and HCOOH more or
less follow the overall DME consumption but HCHO
appears to peak at slightly lower [NO]0. Formation of
CH3OCHO increases in importance with increasing
[NO]0 and is the dominant product once [NO]0 is high
enough to inhibit DME consumption.
OCH2OCHO ! HCOOH ‡ HCO
ꢀ10b†
At 440°C, the e€ect of NO addition is to promote
DME oxidation at all levels of addition although the
The mechanism of Curran et al. (1998) invokes the
species hydroperoxymethyl formate (HPMF,
incremental e€ect beyond ‰NOŠ0=‰DMEŠ ꢁ 1 is negli-
0
gible (Fig. 6). Complete conversion of NO to NO2 is
maintained to higher values of [NO]0 than is seen at
320°C but signi®cant NO is seen in the products at the
highest levels of NO addition. The e€ect of [NO]0 on the
formation of CO re¯ects the increased consumption of
HO2CH2OCHO). At 300°C, the time constant for the
decomposition of this species according to the kinetics
proposed by Curran et al. (1998) is of the order of 10 s,
which is relatively long compared with the reactor resi-
dence time of about 4 s in the present experiments. The

588
I. Liu et al. / Chemosphere 42 (2001) 583±589
chemical kinetic model of Curran et al. (1998) therefore
predicts a substantial accumulation of HPMF.
concern. The USEPA Hazard Summary for formalde-
hyde concludes that it is a probable human carcinogen
of medium carcinogenic hazard (USEPA, 1998); an oc-
cupational exposure limit of 0.3 ppm is recommended.
Toxicity of formic acid is believed to be lower than that
of formaldehyde; it is not classi®ed as an ``air toxic'' by
the USEPA but it is included in the US Toxics Release
Inventory. The occupational exposure limit for formic
acid is typically given as 5 ppm; for methyl formate, the
corresponding value is 100 ppm.
Thamm et al. (1996) have synthesised HPMF
and published its IR spectrum. The species is relatively
stable, with a half life of about 35 min at room temper-
ature and is therefore expected to be amenable to FTIR
detection in the present work also. Careful examination
of the FTIR spectra obtained in this work has failed to
identify any peaks which could be attributed to HPMF
and it is concluded that the concentration of this species
does not exceed 10 ppm in any of the samples studied.
Losses of HPMF on quenching of the reactor prod-
ucts are unlikely to be any greater than those for formic
acid and it is concluded that HPMF does not accumu-
late under the present reaction conditions to the extent
predicted by the model of Curran et al. (1998). While
this discrepancy does not necessarily invalidate the
model, any HPMF formed in this system must disappear
faster than predicted. Either, HPMF is less stable ther-
mally than assumed or that there are other reactions of
HPMF (most likely with OH) which promote the re-
moval of this species.
Formic acid also already plays an important role in
determining precipitation acidity, the acidity of fog and
dew, and the dry deposition ¯ux of acids in urban areas
(Nolte et al., 1997). In addition, formic acid has been re-
ported to have some signi®cant deleterious e€ects on
materials, leading to damage to sandstone, lead (including
lead alloys such as pewter and bronze), shells and fossils.
5. Conclusions
The low-temperature oxidation of DME produces
formic acid as a signi®cant product. In the presence of
NO, DME ignition also produces methyl formate
(CH3OCHO). Qualitatively, these results can be under-
stood in terms of conventional low-temperature ignition
of hydrocarbons. However, recently published chemi-
cal±kinetic models of the low-temperature ignition of
DME do not predict the formation of either of these
formate species. By contrast, the existing models do
suggest that signi®cant quantities of hydroperoxymethyl
formate should be formed under ignition conditions but
this was not observed in the experiments. Further work
is needed to resolve these discrepancies.
The reaction of DME under conditions designed to
simulate atmospheric oxidation leads to substantial
formation of methyl formate, CH3OCHO (Jenkin et al.,
1993), but this species is not detected in this work in the
absence of NO. The internal H-atom shift (reaction (6))
1
has an activation energy of, typically, 80 kJ mol
(Griths and Barnard, 1995; Curran et al., 1998) so this
path is not available at room temperature: under these
circumstances, the self reaction of the accumulating
peroxy species occurs to give rise to the formate (Jenkin
et al., 1993). The very low partial equilibrium concen-
trations of CH3OCH2OO at the higher temperatures of
the present study mitigate against the self-reaction of
these species.
Since both formic acid and methyl formate have
potentially serious harmful health e€ects, these obser-
vations could have signi®cant implications for the pro-
posed use of DME as a diesel fuel.
Once NO is added to the system, substantial quan-
tities of CH3OCHO are now found at the lowest tem-
peratures in the present system. The ability of NO to
promote CH3OCHO formation in atmospheric condi-
tions has been noted previously (Japar et al., 1990) and
the same reaction is likely to be responsible in the pre-
sent system also
Acknowledgements
This work has been supported by the Australian
Research Council.
CH3OCH2OO ‡ NO ! CH3OCH2O ‡ NO2
ꢀ2†
The decomposition of CH3OCH2O to give
CH3OCHO + H is approximately thermoneutral and
probably occurs without signi®cant activation barrier.
The reaction with O2 (reaction (3)) may also be respon-
sible. The conversion of NO to NO2 is also due to reaction
(2) and to reaction with other peroxy species such as HO2.
Even though the formic acid and methyl formate may
be important intermediate products only in the region of
low-temperature ignition of DME, potential emissions
of these species from DME-fuelled vehicles may be of
References
Alzueta, M.U., Muro, J., Bilbao, R., Glarborg, P., 1999.
Oxidation of dimethyl ether and its interaction with
nitrogen oxides. Israel J. Chem. 39 (1), 73±86.
Bowman, F.M., Seinfeld, J.H., 1995. Atmospheric chemistry of
alternate fuels and reformulated gasoline components. Prog.
Energy Combust. Sci. 21 (5), 387±417.
Bromly, J.H., Barnes, F.J., Mandyczewsky, R., Edwards, T.J.,
Haynes, B.S., 1992. An experimental investigation of the

I. Liu et al. / Chemosphere 42 (2001) 583±589
589
mutually sensitised oxidation of nitric oxide and n-butane.
In: Proceedings of the Twenty-Fourth Symposium (Inter-
national) on Combustion, The Combustion Institute, Pitts-
burgh, pp. 899±907.
Hay, J.M., Hessam, K., 1971. Oxidation of gaseous formalde-
hyde. Combust. Flame 16 (3), 237±242.
Hjuler, K., Glarborg, P., Dam-Johansen, K., 1995. Mutually
promoted thermal oxidation of nitric oxide and organic
compounds. Indust. Eng. Chem. Res. 34 (5), 1882±1888.
Japar, S.M., Wallington, T.J., Richert, J.F.O., Ball, J.C., 1990.
The atmospheric chemistry of oxygenated fuel additives: t-
butyl alcohol, dimethyl ether, and methyl t-butyl ether. Int.
J. Chem. Kinetics 22 (12), 1257±1269.
Bromly, J.H., Barnes, F.J., Muris, S., You, X., Haynes, B.S.,
1996. Kinetic and thermodynamic sensitivity analysis of the
NO-sensitised oxidation of methane. Combust. Sci. Tech-
nol. 115 (3), 259±297.
Curran, H.J., Pitz, W.J., Westbrook, C.K., Dagaut, P., Boett-
ner, J.-C., Cathonnet, M., 1998. A wide range modeling
study of dimethyl ether oxidation. Int. J. Chem. Kinetics 30
(2), 229±241.
Jenkin, M.E., Hayman, G.D., Wallington, T.J., Hurley, M.D.,
Ball, J.C., Nielsen, O.J., Ellermann, T., 1993. Kinetic and
mechanistic study of the self reaction of CH3OCH2O2
radicals at room temperature. J. Phys. Chem. 97 (45),
11712±11723.
Dagaut, P., Boettner, J.-C., Cathonnet, M., 1996. Chemical
kinetic study of dimethylether oxidation in a jet stirred
reactor from 1 to 10 atm: experiments and kinetic modeling.
In: Proceedings of the 26th Symposium (International) on
Combustion, The Combustion Institute, Pittsburgh,
pp. 627±632.
Maricq, M.M., Szente, J.J., Hybl, J.D., 1997. Kinetic Studies of
the oxidation of dimethyl ether and its chain reaction with
Cl2. J. Phys. Chem. A 101 (28), 5155±5167.
Nolte, C.G., Solomon, P.A., Fall, T., Salmon, L.G., Cass,
G.R., 1997. Seasonal and spatial characteristics of formic
and acetic acids concentrations in the southern california
atmosphere. Environ. Sci. Technol. 31 (9), 2547±2553.
Sehested, J., Mogelberg, T., Wallington, T.J., Kaiser, E.W.,
Nielsen, O.J., 1996. Dimethyl ether oxidation: kinetics and
mechanism of the CH3OCH2 ‡ O2 reaction at 296 K and
0.38±940 Torr total pressure. J. Phys. Chem. 100 (43),
17218±17225.
Dagaut, P., Daly, C., Simmie, J.M., Cathonnet, M., 1998. The
oxidation and ignition of dimethyl ether from low to high
temperature (550±1600 K): experiments and kinetic model-
ing. In: Proceedings of the 27th Symposium (International)
on Combustion, The Combustion Institute, pp. 361±370.
Doughty, A., Barnes, F.J., Bromly, J.H., Haynes, B.S., 1996.
The mutually sensitised oxidation of ethylene and NO: an
experimental and kinetic modelling study. In: Proceedings
of the 26th Symposium (International) on Combustion, The
Combustion Institute, Pittsburgh, pp. 590±596.
Sehested, J., Sehested, K., Platz, J., Egsgaard, H., Nielsen, O.J.,
1997. Oxidation of dimethyl ether: absolute rate constants
for the self-reaction of CH3OCH2 radicals, the reaction of
CH3OCH2 radicals with O2, and the thermal decomposition
of CH3OCH2 radicals. Int. J. Chem. Kinetics 29 (8),
627±636.
Fleisch, T., McCarthy, C., Basu, A., Udovich, C., Charbon-
neau, P., Slodowske, W., Mikkelsen, S.-E., McCandless, J.,
1995. New clean diesel technology: demonstration of ULEV
emissions on a Navistar diesel engine fueled with dimethyl
ether. SAE Special Publications 1129, SAE, pp. 39±50.
Grith, D.W.T., 1996. Synthetic calibration and quantitative
analysis of gas-phase FT-IR spectra. Appl. Spectroscopy 50
(1), 59±70.
Thamm, J., Wol€, S., Becker, K.H., 1996. Proof of the
formation of hydroperoxymethyl formate in the ozonolysis
of ethene: synthesis and FT-IR spectra of the authentic
compound. Chem. Phys. Lett. 258 (1/2), 155±158.
USEPA, 1998. Hazard Summary, http://www.epa.gov/ttn/uatw/
hlthef/formalde.html.
Griths, J.F., Barnard, J.A., 1995. Flame and Combustion,
third ed. Blackie Academic and Professional, London,
pp. 129±138.