close

Вход

Забыли?

вход по аккаунту

?

Mechanisms of Photochemical Air Pollution.

код для вставкиСкачать
Volume 14 Number 1
January 1975
Pages 1- 74
International Edition in English
Mechanisms of Photochemical Air Pollution
By J. N. Pitts, Jr. and B. J. Finlayson[']
Dedicated to Professors Philip A. Leighton and Francis E. Blacet
Selected aspects of the chemistry of photochemical air pollution is discussed and some important,
unresolved problems dilineated. The reactive species considered include NO,, 0,, O(3P), O( 'D),
02('AJ,
OH and HO,. Both the kinetics and mechanisms of the reactions constituting the major
tropospheric sources and sinks of these species are treated where available. The application of
this information in both computer and smog chamber simulations of photochemical smog is
discussed.
1. Introduction
Photochemical air pollution was first overtly recognized in
the Los Angeles Basin in the 1940's from the crop damage
it caused[']. This new type of smog was characterized by
the formation of oxidants[*' in contrast to the well-known
"London Smog" which contained primarily SO, and particul a t e ~ ' ~Haagen-Smit
].
and co-workersr4-'I demonstrated that
a mixture of cracked gasoline or olefins and ozone acting
on crops reproduced the smog damage, and showed that
ozone was produced when a mixture of nitrogen dioxide and
individual organics (olefins, alkanes, alcohols, carbonyl compounds, acids) or dilute auto exhaust was irradiated.
Since then these experiments have been reproduced under
a variety of conditions. Figure 1 shows a typical time dependence['-''] of the reactants and major products when a mixture of nitrogen oxides and propylene in the part-per-million
(ppm) concentration range are irradiated in air with ultraviolet
light (h2290nm)[39"I. Initially, NO is rapidly converted to
NO,. Simultaneously, the olefm concentration decreases and
acetaldehyde appears. When the NO concentration is sufticiently low, 0, and peracetyl nitrate (CH,CO,NOJ build
up while the NO, and C,H, concentrations fall.
['I
Prof. Dr. N. Pitts, Jr. and Dr. B. J. Finlayson[-l
Department of Chemistry and
Statewide Air Pollution Research Center
University of California
Riverside, Calif. 92502 (USA)
p*]Present address:
Department of Chemistry,
California State University
Fullerton Calif. 92634 (USA)
Angew. Chem internat. Edit.
Vol. 14 (1972) J No. J
-
t Cmin 1
113211
Fig. 1 . Typical concentration-time profiles [8-lo] for the reactants and
major products in the photooxidation of propylene in the presence of NO
and NO, (after [lo]).
In this system NO, is the major light-absorbing species[3!
At wavelengths less than 430nm NO, dissociates forming
NO and ground state oxygen atoms, O('P), (see Section 2):
NO,
h <430nm
NO
+ o(3~)
(1)
In air this is rapidly followed by:
*I
0
+ 0, + M
+ NO
0,
+
--t
+M
0,
NO,
+ 0,
ky’=2.1 x 1 0 8 ( M = N 2 )
k, = (1.1 0.2) x 10’
(2) [12, 131
(3)
[I41
was shownrzo1to be fast, may now assume an important
role under appropriate conditions. Some organics, such as
formaldehyde, may act in a similar manner[”’:
In the presence of propylene, we should include:
O+C,H6
0, + C,H6
4
--t
OH + HCHO 4 H,O + HCO
k , = (8.4 f 2.1) x 10’
HCO + 0, + HO, + CO
k , , = ( 3 4 f 0.7) x 10’
HO,+NO 4 HO+NO,
products
products
The behavior shown in Figure 1 has challenged chemists
to explain the following apparent anomalies. Starting with
large quantities of NO and small amounts of NO,, we see
in the initial stages of the reaction an increase in NO,, the
species which we know is being lost photochemically by reaction (1). Secondly, the rate of olefin disappearance is much
greater than that due solely to its reactions with 0 and 0,,
as shown in Figure 2[’01. These anomalies imply that a chain
reaction must be involved both in the NO-NO, conversion
and in the olefin loss.
50
100
.
150
t [rninlFig. 2. Comparison of the experimentally determined rate u of propylene
loss to that calculated from its reaction with 0 and 0, [S-lo]
(after
[101).
The behavior described above is characteristic of photochemical smog formation involving any reactive hydrocarbon, singly
or in complex mixtures such as auto exhaust. In 1961, Leighspeculated that radicals such as R, RO,, RO, OH, HO,,
and H might be involved. Further work since then has strengthened and clarified these ideas. One major contribution was
the suggestion[’ 5 . I 6 l that CO, in conjunction with hydroxyl
(OH) and hydroperoxyl radicals (HO,) might participate in
the NO to NO, conversion :
OH
+ CO
+
-t CO,
H
9.8 x 10’
H + O z + M --t H 0 2 + M
k , = (2.13 0.4) x 10” (M = Nz)
HO, + NO 4 NO,
OH
k , = (1.8 f factor of 3) x 10’
k6
Thus, any source of H, HO,, or OH will contribute to the
rapid oxidation of NO to NO, and, as we shall see, to the
enhanced rate of olefin consumption.
Understanding the detailed course of the reactions involved
in complex smog mixtures such as that in Figure 1 is the
immediate goal of atmospheric chemists today. This necessitates an intimate knowledge of the kinetics and mechanisms
(including intermediates and products) of all elementary reactions involved. Only then can we accurately and reliably predict the time dependence of the pollutant concentrations in
such simulated smog experiments. Of course, even this mammoth task is only a step towards the ultimate aim-predicting
the same time dependence in the real atmosphere where fresh
pollutants are constantly being added, meteorology is changing, and some of yesterday’s “aged” smog is also present
when the sun rises to initiate the reactions today. In order
to meet even the initial goal, however, we lack kinetic and
mechanistic information in many crucial areas, information
which must be obtained by careful, quantitative research in
a variety of systems ranging from the torr to part-per-billion
(ppb) concentration ranges. This provides an unparalleled
opportunity for chemists to apply their expertise to a societal
problem facing all of us.
In this article, we briefly summarize current knowledge of
selected areas of photochemical air pollution, and in doing
so, delineate some of the present and future challenges to
the chemist. This is not intended to be a comprehensive review
but rather an examination of certain interesting aspects of
smog chemistry which we hope will put the subject in perspective. We apologize in advance to the many scientists whose
work is not included, or discussed in the detail it deserves,
due to the limitations of space.
2. Photolysis of Nitrogen Dioxide (NO,)
NO, is the major light-absorbing species in the solar actinic
region (h>290nm) and is an important (but not the sole)
initiator of photochemical smog formation. It has been establ i ~ h e d [that
~ I the primary process on absorption of wavelengths
less than ~ 3 9 nm
8 is dissociation, with a primary quantum
yield of approximately unity:
=
+
NO,(X~A,)
2 NO(XW +
o(3~)
(1)
Thus, carbon monoxide, which was assumed to be unreactive
in photochemical air pollution until the rate of reaction (6)
This is consistent with the NO, bond dissociation energy
of 3.1 15eV ( h= 398 nm)[241. At wavelengths greater than
430nm, there is no O(,P) p r o d ~ c t i o n ~ ~only
- ~ ~photo-~~~;
physical processes occur.
[*] Rate constants throughout are given in I-mol-s units at 300°K.
Bond rupture also occurs, however, at wavelengths between
398 and 430 nmr3.25.26.291, as confirmed by a recent study[28].
Anguw. Chem. internat. Edit.
1 Vol. 14 ( 1 9 7 5 ) N 0. I
In this wavelength region the photon is not suficiently energetic by itself to cause dissociation, and the excess energy must
351.
come from the rotational energy'".
Recently, Jones and Bayes['*' measured relative quantum
yields for NO formation at exciting wavelengths from 295
to 445nm in 5 or 1Onm intervals using a photoionization
mass spectrometer. This technique, which has proven to be
powerful and will undoubtedly play a large role in kinetic
and mechanistic studies in the future, uses photon absorption
to ionize the species entering a quadrupole mass spectrometer.
The advantages over ionization by conventional electron
impact include (i) the possibility of selective ionization (through
the choice of ionizing wavelength) of small quantities of species
in the presence of large amounts of compounds which would
otherwise interfere, (ii) minimization of fragmentation processes, hence greatly simplifying interpretation of the mass
peaks, and (iii) partial confirmation of identification through
the use of resonance lamps of varying photon energy'28.301.
were put on an absolute
The quantum yields for
basis using the results of previous work at 313 and 366nm.
Since reaction (1) is rapidly followed by
0
+ NO,
-
NO
+ O2
(11)
(DNOis anticipated to be 2.0 at h<398nm in the absence
of competing processes.
0
s ~ o o . o0
*,
00
0
A
q
\
s t ~ d i e d ~quantitatively,
~l;
however, the calculated NO quantum yields are lower[281than those determined experimentally.
Further studies are necessary in order to resolve this discrepancy between theory and experiment.
3. The Role of Ozone in Photochemical Air Pollution
In air, the ground state oxygen atom, O(3P), produced in
the photolysis of NO,, is removed primarily by reaction (2)
with 0 2 . At the present time, this is the only reaction known
to produce O3 although alternate reactions such as that of
the Criegee diradical (see Section 3.2) with 0, have been
suggested[361.
RCHOO. +
o2
+
RCHO
+ o3
(12)
3.1. Kinetics of the Ozone-Olefin Reaction
Table 1 gives typical room temperature rate constants for
the reactions of O 3 with some species normally present in
polluted urban atmospheres. Using these rates and typical
ambient pollutant concentrations, one calculates that the only
important loss processes of O 3occur by reaction with olefins,
NO, and to a lesser extent, NOz. The rapid reaction (3)
of 0 3 with NO is the reason that 0 3 does not begin to
accumulate (Figure 1) until the NO concentration has decreased to a low value. This may also at least partially explain
why oxidant levels in downtown Los Angeles have dropped
as the NO emissions have risen; concurrently, ozone concentrations have increased in communities downwind due to
the subsequent photolysis of the NO2 formed by reaction
(3)[41-431.
I
lo
I
Table 1. Selected room temperature rate constants for some ozone reactions.
I
I
Compound
I
I
I
i,
0
'00
Q ,
Fig. 3. Wavelength dependence of the quantum yield [28] for NO formation
in the photolysis of NO,; o from ref. [28]; A from ref. [3, 261. Dotted
line gives the dissociation energy of NO, (3.115eV) [24J Solid line is the
least squares fit to the data of ref. [28] (after [ZS]).
Figure 3 gives the results of this investigation[281 as well
as those of earlier s t ~ d i e s [ ~ . It
~ ~ !is seen that for
295 < k < 398 nm, @ is close to 2 but decreases slightly with
increasing wavelength. Junes and Bayes[z81suggested that
the competing process which accounts for z3 % of the reaction
at 31 3 nm and % 8 at 366 nm may be energy transfer to
0 2 . Thus these investigator^[^'.^^^ observed the production
of electronically excited OZ(a'A,) when mixtures of NOz and
0 2 were irradiated in the region 330<h<590nm. Frankiewicz
and Berry133*
341 also observed the production of small quantities of O,(b'X:)
in the same system.
As mentioned above, the observed dissociation of NO, at
h > 398 nm can be qualitatively explained by the contribution
of the rotational energy, in agreement with recent spectroscopic
Angrw. Chum. intrmat. Edit.
kxi0-3
Ref.
[1 mol-' S K ' ]
16
1 Vol. 14 ( 1 9 7 5 ) 1 No. I
Ethylene
Propylene
Isobutene
trans-2-Butene
(CH 3)2C=C HCH 3
(CH>)zC=C(CH3)2
Toluene
HC=H
CHKHO
CHI
co
NO
NO 2
0.93 f0.09
7.5k0.6
8.2
165f 14
296? 10
906f48
7.zX10-3
52 x lo-'
2oX 1 0 - 3
50.72 10-3
10.6 x
(1.1 + 0 . 2 ) ~104
19f3
At low NO concentrations O 3 reacts primarily with olefins
a t rates of z 103-105 1mol- s- While the rate constants
of Table 1 are generally the most recently determined ones,
there are significant disagreements in the literature values,
particularly for olefins containing an internal double bond.
It now appears that secondary attack of intermediates on
the reactants makes the experimentally determined rate constant sensitive to the reaction conditions, particularly the 0,
c ~ n c e n t r a t i o n ' 4~4!~ . Additionally, the rate of formation of
the major carbonyl products may be nonlinear[451, at least
under certain conditions, and hence those rate studies based
solely on product formation may be in error. These factors
have only recently come to light and indicate the need for
'.
3
(1-hexene, 2-pentene, and 3-heptene) where diradical decomposition is not expected to be extensive. Presumably this arises
by recombination of the diradical and carbonyl fragments
in a manner analogous to the liquid phase [reaction (15)].
further kinetic studies over a wide variety of experimental
conditions.
3.2 Mechanism of Gas Base Ozone-Olefin Reactions
Until recently, the mechanism of the ozone-olefin reactions
in the gas phase had generally been interpreted in terms
of a modified liquid phase Criegee m e c h a n i ~ r n [ ~ ~ -The
~*!
initial electrophilic attack of ozone on the olefin is thought
to produce a molozonide, or primary ozonide, which cleaves
to the Criegee “zwitterion” (more likely a diradical in the
gas phase) and a carbonyl compound [reactions (13a) and
(13b)] :
molozonide
(prim. ozonide)
\ a
c-0-0’ +
i
R R‘3
O
X0XR4(15)
/”
prim. ozonide
\
( 1 3 ~
According to Scott, Hanst, and c o - w o r k e r ~ [501,
~ ~the
. diradical
then decomposes, rearranges or reacts further. The extent
of decomposition will be determined primarily by the size
of the diradical since larger molecules will have a greater
number of vibrational degrees of freedom over which to dissipate the excess energy without decomposing. For example,
the Criegee diradical CH,CHOO from the propylene reaction may decomposer49.501 as follows:
+ R3R‘C=0
R’ o-0 R3
R
R’R~C=O
R4’
R’R’dOO.
Ra
R3R4eOO- + R’RaC=O
/*
sec. ozonide
However, the formation of such ozonides has not been reported
using alternate techniques such as GC. Such confirmatory
studies are, in the authors’ opinion, desirable.
Subsequent studies using a variety of techniques over a wide
range of experimental conditions have yielded results which
are not easily explained by this mechanism, however. These
include (i) the formation of a wide variety of “abnormal”
ozonolysis products such as peroxides; aldehydes, and ketones
one and two carbons shorter than the major carbonyl product;
and carbonyl compounds having the same chain length as
the reactant 01efint5 - 541$ (ii) non-unit stoichiometry[38.52*55.561, (iii) the formation of radical intermediates at
low
including significant yields of H
atoms[44.57.59], and (iv) the production of chemiluminescence[44.57.59-63]
CH,--CH-O-O.
(a) H,O
+ CH,=C=O
2 CH,OH + CO
0CH.,+CO,
All of these products were observed in the propylene-ozone
reaction[49’501.
On the other hand, secondary ozonide formation was observed
by infrared s p e c t r o s ~ o p yfor
~ the
~~~
longer chain hydrocarbons
In order to explain these observations, ONeal and Blumstein[64]have proposed that, in addition to the Criegee path,
u, p, and y abstractions of the primary ozonide can occur.
For example, for the 1-butene reaction with 0 3 , the following
series of reactions are postulated; the relative importance
of each of these paths under various experimental conditions
can be calculated from the thermochemistry and unimolecular
theory. The A H o values are -32.9 for reaction (16), -74.3
for (17a), -65 for (18), -80 for (19), and <4.4kcal/mol for
(20a).
Initial addition
‘90
0-0
CH,eH-AHCHaOH
C H 3 H CHzOH
H H
IP-H
(17a)
?OH
CH3CHaCH-CHO \-H
qO’b
7 - y
EHzbaHCHaOH-
H
zH
Q (19) ~
Ha
CH3CHzCH- &HZ
C H , C H ~ ~ H O O .+ HCHO
CH,CH,CHO
4
+
H~HOO.
(20b)
Angew. Chem internat. Edit.
1 Vol. 14 (1975) 1 No. 1
~
Chemiluminescence is then rationalized by the decomposition
of the dioxetanes formed by j3-hydrogen abstraction[44.64, 651,
leading to electronically excited species.
Reactions (16)-(20) are highly exothermic and hence will
result in the formation of excited products. Decomposition
of these energetic species will occur in competition with their
collisional stabilization and hence be pressure dependent. For
example, excited hydroperoxides formed by a-hydrogen
abstraction, as in (17), are known["] to cleave at the 0-OH
bond if sufficient energy is available [reaction (21)].
Because of the uncertainties regarding the reaction mechanism,
the precise role of ozone-olefin reactions in photochemical
smog formation still remains obscure. One possibility is the
production of H atoms followed by their reaction with 0 2
to form H02[441.Furthermore, if radicals such as HCO which
are formed at 2t0rr[~*lare also produced at atmospheric
pressure, they too may participate in the NO to NO2 conversion by formation of HO,, by reaction (10). Finally, it is
interesting to speculate that the u-carbonyl hydroperoxide
may be the phytotoxicant produced in ozone-olefin reactionS[7i.721,
3.3. Effect of Sulfur Dioxide on Ozone-Oletin Reactions
A novel feature of the O'Neal-Blurnstein proposal is the postulated rearrangement of the Criegee diradical under certain
conditions to theexcited acid which may also decompose[64-671
[reaction (22)l.
When SO2 is added to a mixture of ozone and olefin in
the dark or to dilute auto exhaust which is subsequently
irradiated, it is rapidly oxidized by some unidentified reaction
intermediate or product to sulfuric acid aerosol[45.7 3 - 7 5 1 . An
ozone-olefin adductL4'],the Criegee d i r a d i ~ a l [ and
~ ~ ] H02[441
,
have all been suggested as the oxidizing species [reactions
(25)-(27)J.
-
[RCH=CHK*03] + SO2
R'CHOO.
COz + H'
+
soz
HOz + SO2
2 RCHO + S
0
3
so,
-+
RCHO +
.-+
H O + SO3
(25)
b
Has04
(26)
(27)
This may explain the production of H a t o r n ~ [ ~ ~ .and
' ~ . ' ~ ~ While further detailed studies are badly needed, this important
also be consistent with the proposals of Scott, Hanst, and
finding may explain the high rates of oxidation of SO2 in
coworkers[49, regarding the modes of decomposition of
areas of photochemical smog[z11and would suggest that the
the diradical CH3CHOO. (see above). Acetic acid is known
use of high sulfur fuels in such areas be discouraged[431because
to decompose photochemically to C H 4 and C02I6'' and therof the apparent toxicity of particulate
and its role
mally to CH *=C=O
H20["I. Ketene has been observed
in visibility reduction.
in significant yields from the reaction of cis-2-butene with
ozone both at lowf5*3701 and at highL7'] total pressures, while
4. Reactions of O(3P)with Olefins and Arenes
CH,COOH was not detected['*. 701.
+
At high pressures and for long chain diradicals the acid from
the Criegee cleavage may be stabilized, in which case the
final products are the same as those from the decomposition
of the a-carbonyl hydroperoxide [reactions (23) and (24)].
?OH
CH~CHZCH-CHO + CH3CHzCHO + HCOOH
4.1. Kinetics of Reactions of O(3P) with Olefins and Arenes
(23)
The only means of differentiating between these two paths
is identification of the u-carbonyl hydroperoxide which is
an intermediate in the a-hydrogen abstraction path but not
in the Criegee path.
While this theory of ozone-olefin reactions is attractive, further
experimental validation is needed before the theory, or selected
aspects thereof, can be accepted. It is evident that much more
work is needed on the intermediates (including radical species),
stoichiometry, effects of oxygen, total pressure, and reactant
concentrations using a variety of techniques in order to establish confidently the mechanisms and the bimolecular rate
constants of these complex reactions.
Angrw Chem. internat. hdit. j Vol. 14 ( 1 9 7 5 )
1 No. I
At typical ambient pollutant levels O(3P) reacts to a small
extent with hydrocarbons, in addition to its primary loss
by reaction (2) with 02.
Table 2 gives the values of selected rate constants for the
reactions of O(3P)with some olefins and arenes. These rates,
which show an electrophilic
represent addition to
the double bond. While it was suggested"*] that abstraction
may occur z 15 of the time at 300°K in the O(3P)-l-butene
reaction, O H was not detected[791as an intermediate in this
reaction, supporting previous assumptions that hydrogen
abstraction is too slow to compete with addition at room
temperature.
4.2. Mechanism of the 0(3P)-Olefin Reaction
The results of a definitive series of studies of the 0(3P)-olefin
reactions led Cuetanouic and c o - w o r k e r ~ ~to' ~postulate
~
the
following reaction mechanism, illustrated by the reaction of
O(3P) with cis-2-pentene [reactions (28)-(31)].
5
Table 2. Typical rate constants k at 298") for the reaction of O('P) with olefins [165] and arenes [77].
Olefin
Arene
kxiO
[Imol I s - ' ]
Ethylene
Propylene
1-Butene
cis-2-Butene
2-Methyl-2-butene
2.3-Dimet hyl-2-butene
4.3: 0.5
20 I 1.7
24 j 3.7
92 1 1 5
313 3 30
425 2 16
Benzene
Toluene
o-Xylene
m-Xylene
p-Xytene
1,2,3-Trimethylbenzene
1,2,4-Trimethylbenzene
1,3,5-Trimetbylbenrene
0.144 & 0.02
0.45 kO.045
t.05 51.11
2.12 k0.21
1.09 kO.11
6.9 k0.7
6.0 k0.6
16.8 22.0
J
U
According to this scheme:
1) O(3P)adds to the double bond, primarily to the less substituted carbon atom, forming a triplet diradical [reaction (2811.
2) The diradical is sufficiently long-lived that some rotation
about the original double bond occurs since both cis and
trans epoxides are formed starting from one geometric isomer
of the olefin [reaction (29)].
3) Ring closure to the epoxide must be competitive with rotation since the ratio of cis to trans epoxides formed is not
the same for the reaction of the cis and trans olefins respectively.
4) The epoxide formed contains x 90kcal/mol excess energy
and hence will decompose [reaction (29a)l unless first collisionally stabilized [reaction (29b)f.
5 ) The diradical may also undergo an internal migration of
an H atom from the carbon to which the oxygen is attached
to the adjacent carbon to form methyl n-propyl ketone containing ~ 1 1 5 k c a lexcess energy; the excited ketone will then
decompose [reaction (3Oa)lif not collisionally stabilized [reaction (3Ob)l.
6 ) The alkyl group attached to the carbon attacked by the
Ot3P)may also migrate to the neighboring carbon, ultimately
forming excited 2-methylbutanal which can decompose [reaction (31 a)] or be stabilized by collisions [reaction (31 b)].
7) The migration of the alkyl group is partly internal and
partly external in nature, as indicated by the formation of
small amounts of CHh and ClHe.
6
8) This splitting off ofan alkyl radical from the triplet diradical,
its third available reaction mode, is not quenched by increasing
pressures and hence is referred to as the "pressure independent
fragmentation.''
This reaction scheme, postulated on the basis of the observed
stable products, has been confirmed recently by the detection
of the intermediate radical species using photoionization mass
spectrometry[80-831. These studies have also elucidated minor
reaction paths which occur in addition to those described
above.
One of the major unsolved mechanistic problems of the q 3 P )
reaction with olefins, relevant to photochemical smog, is the
effect of excess molecular oxygen at atmospheric pressure.
Such studies are hampered by the possibility of ozone formation via reaction (2) since ozone also reacts with olefins to
produce a wide variety of products including carbonyl compounds.
In one study[64'where NO2 was used as the source of 0f3P)
and increasing amounts of 0 2 were added to the cis-2-pentene
reaction, significant quantities of acetaldehyde and propanal
were formed, the epoxide yield decreased and the yields of
methyl propyl ketoneand 2-methylbutanai remained constant.
Under these conditions O 3 formation should be negligible
and it therefore appears that Oz must interact with some
of the intermediates. The triplet diradical is probably sufficiently short-lived[76]so that no collisions occur before it
undergoes further reaction. Since the yields of the addition
carbonyl compounds are also unaffected by increasing pressures of 02[841,
it appears that the most likely interaction
of 0 2 is with the excited epoxides. However, until there is
much more data available on the effect of 0 2 on these reactions
and on their intermediates, a detailed mechanistic discussion
is not possible. Such studies are particularly important in
that some monofunctional epoxides show carcinogenic activity
in mice and rats[851.Such epoxides include 3,4-epoxy-l-butene
and styrene oxide which one might envision as being formed
from oxygen atom reactions with 1,3-butadiene and styrene,
respectively. It must be stressed however, that the production
of such epoxides under ambient air conditions and their presence in polluted urban atmospheres at detectable levels has
not been confirmed.
4.3. Mechanism of the O(3P)-AreneReaction
In 1961 Cvetanovic and c o - w o r k e r ~ reported
~ ~ ~ . ~ that
~ ~ the
volatile products of the reactions of O(3P) with benzene and
toluene are water and CO, and phenol (from C6H6) and
cresols (from C7H8), respectively, with the major product
being nonvolatile polymeric compounds which deposited on
the walls of the cell. This has recently been confirmed[88.921.
The polymeric material is thought to contain such functional
groups as --CHO[86.891,-OH, and C-O-Cf891 and to
be aliphatic in nature[86! While the formation of this
"polymer" has repeatedly haunted researchers, it is interesting
from the point of view of photochemical air pollution in
that significant quantities of aliphatic difunctional compounds
recently have been observed in ambient air particulate matter[90,91!Many of these compounds contain five and six
It is obvious that the reaction of O(,P) with arenes is very
complex and will require the use of highly sophisticated techniques such as photoionization mass spectrometry in order
to clarify the reaction mechanisms. However, with the increasing arene content of gasoline as lead is removed, even higher
levels of these compounds in ambient air are anticipated.
Hence, these reactions may become increasingly important,
particularly if it can be conclusively demonstrated that ring
opening to produce unsaturated aliphatic compounds is a
major reaction pathway.
Although highly speculative at the present time, another possible route to ring opening in photochemical smog may involve
the oxygen-enhanced absorption of light by arenes followed
by their photooxidation to difunctional unsaturated compounds. Thus, in solution the irradiation of oxygen-saturated
benzene led to the formation of both six and twelve-carbon
conjugated dialdehyde~[~~!
5. Reactions of Electronically Excited O('D) Atoms
Electronically excited O(ID) atoms containing 45.4 kcal/mol
excitation[94' are produced in polluted urban atmospheres
by the photolysis of O3 at wavelengths less than 308nm[951.
While the major tropospheric removal process of O('D) is
physical quenching to O(,P) primarily by 0 2 and Nz, water
is typically present at concentrations of = 1-3 % and hence
can account for chemical quenching of ~ 9 of
% the O('D)
atoms f ~ r m e d [ ~ ~ , ~ ~ ] :
O('D)
&"-
.CH=CH-CH=CH-CH=C,
P
(33)
0'
carbon atoms and hence we feel may arise from the attack
of reactive radicals such as O(,P) and OH on an aromatic ring,
followed by ring opening and subsequent reactions with 0,,
etc. For example, following the initial addition of O(3P) to
the benzene ringrs6][reaction (32)], ring opening may occur
according to reaction (33)[*']'.
Subsequent reactions of the unsaturated diradical may then
produce the nonvolatile products[86].
Alternatively, the following Scheme (34)has been proposed[861
to account for the observed products :
I
-
+Polymer
CHO
+
co
Such a mechanism cannot be extended to more complex arenes,
as shown for example, by the lack of formation of 2,6-dimethylanisole in the 1,2,3-trimethylbenzenereactiontg2!
Angrw. Chem internat. Edit.
1 Vol. 14 ( 1 9 7 5 ) J N o . 1
+ H20 32 0 H
It is estimated that more than 90% of the reaction produces
hydroxyl radicals (path a)1981.
Hence, at high ozone concentrations in severe photochemical smog, this reaction may be
an important source of O H radicals (see Section 7.1).
Recently, it has been shown that, in contrast to O(,P) whose
rates of reaction with chlorofluoroalkanes are very
O('D) atoms react at almost collision frequency with a variety
of these compoundsf'OO~'O'l.
Thus, the rate constants for its
reactions with CFCl,, CF,CI,, CHF,Cl, and CF,CICF,CI are
3 . 5 lo",
~
3.2 x lo", 2.1 x lo", and 2 . 0 101llmol-ls-l,
~
respectively. Major products with CFC13 and CF2CI2 are
COFCl and COF2, respectively. Probably, chlorine is also
formed either as Clz in a molecular elimination process or
as a recombination product of C1 atoms. This is currently
being investigated.
Since chlorofluoroalkanes are commonly used as aerosol propellants and are released into the atmosphere in great quantities, their atmospheric sinks are of considerable interest. In
the troposphere, the concentration of O('D) is sufficiently
low that reaction with chlorofluoroalkanes is not a significant
loss process, and, as indicated above, other tropospheric reactions of these compounds are slow. In the stratosphere, however, the O('D) concentrations are much
and reaction with chlorofluoroalkanes that may have undergone vertical transport from the troposphere could prove to be signifi7
Recent calculations by Firestone and Calvert[' 221, however,
cant['OO.'O'l.This is especially true if either Cl2 or C1 are
indicate that we should be cognizant of possible health probformed in the primary attack of O('D) or if COFCl (arising
from CFC13) is photolyzed to produce C1, as seems likely.
lems associated with 0 2 ( 'Ag). They calculate that the half-life
for quenching of O2('Ag)by 0, is sufficiently great ( 9 . 0 ~
Thus Molina and Rowland have recently proposed that chlomin) compared to the period for the inhalation cycle
rine atoms from the photodissociation of chlorofluoroalkanes
by short wavelength solar radiation in the stratosphere['03]
(3.5 x lO-'min), that Oz('Ag)may reach the lung unquenched
could lead to a chain destruction of stratospheric 03[103.1041
a significant portion of the time. Depending on the nature
of the products formed by the interaction of 02(lAg) with
in a manner similar to that suggested for nitric oxide.
the lung tissue, some adverse health effects may arise.
Whether or not either of these proposed mechanisms for
Hence the presence of 02('Ag) in polluted atmospheres is
degradation of chlorofluoroalkanes, direct photodissociation
of concern for several reasons: (i) its calculated lifetime is
or attack by O('D) atoms, actually poses a problem depends
sufficientlygreat that significant quantities may penetrate into
upon several factors: (i) validation of these mechanisms under
the lung and react to form toxic compounds; (ii) it may
simulated stratospheric conditions ; (ii) establishment of presreact with a variety of olefins and arenes[123-1291
and in
ent and predicted future chlorofluoroalkane concentrations
ambient air form such known toxins as hydroperoxin the stratosphere; and (iii) their importance as chlorine
ides[85.129-131] which may then be inhaled; and (iii) it may
atom sources relative to other precursors to C1already existing
be formed in situ by hydrolysis of air pollutants such as
in the stratosphere. Clearly, research in these areas is vitally
peracetyl nitrate" 641.
needed.
Further work seems warranted to establish the ambient levels
of Oz('Ag),to clarify its reactions with organics in ambient
6. Singlet Molecular Oxygen, O#A,)
air and with biological components, and to establish the effects
of the products of these reactions on human organisms.
Leightonl3] first considered the electronically excited states
of 0 2 as potential atmospheric oxidants, but he concluded
that their production by the direct absorption of sunlight
7. The Hydroxyl Radical (OH): Kinetics and
~ ~ ] showed that
was too small to be significant. B a y e ~ " then
MeCbanisms
if the 0 2 absorption coefficients are corrected for collisional
line broadening, excited 02('Ag) may be formed in high enough
The suggestion that the OH radical is an important chain
concentrations and have a sufficiently long lifetime to react
carrier in photochemical smog formation[3,15. ' has sparked
with simple olefins in the atmosphere, for example, according
much interest in both its kinetics and reaction mechanisms.
to reaction (36)[lo6.'07!
This emphasis has been given added impetus recently by
the detection of OH in ambient air at daytime concentrations
of the order of 108/cm3using a tunable UV laser['321.
Subsequently, it was proposed['08. lo91 that energy transfer
from triplet excited organic pollutants to ground state oxygen
might yield significant concentrations of 0 2 ( 'Ag). Additional
sources of 02(lAg) in polluted urban atmospheres[' lo.llll
include energy transfer to 0, from N0,[3'-34' (see Section
exothermic chemical reactions
2), photolysis of 0 3 1 ' '
oxygen-enhanced absorption by
producing 0 2 [ ' "I,
organics" 6],and the hydrolysis of peracetyl nitrate''64!
Quantitative estimates of the yields of 02('Ag) from each
of these processes are difficult. However, a recent estimatet3']
of the 0z('Ag) concentration due to the sum of direct absorp
tion, energy transfer from NO:, and photolysis of 0, is x
PPm.
that 0 2 ( 'Ag) might in part be responIt was thought['08sible for the excess rate of consumption of olefins (Fig. 2)
and the conversion of NO to NOz if its rate constants for
reactions with olefins were sufficiently high, a point that was
at that time unknown. Subsequently, these rate constants
were found to be comparable to those of ozone-olefin reactions,
i. e., approximately five orders of magnitude lower than those
of the corresponding O(jP) reactions[111.1'7-1191. Since the
O(3P) concentration is estimated[31to be ~ I O - ~ p p and
m
the oxidation
if the 0 2 ( lAg) concentration is % 10of olefins by 02('Ag) will not contribute significantly to the
bulk consumption of olefins in polluted urban atmosphere [111.120.1211 (see Table 5).
'-
'
8
7.1. Sources of the OH Radical
As mentioned above, O('D) atoms formed during ozone photolysis react with water to produce 2 OH radicals [reaction
(35a)l. The reaction of HOz with NO [reaction @)I,and
the photolysis of nitrous acid'133' are also sources of OH
in the polluted troposphere.
HONO
+ hv
1<400nm
____*
OH
+ NO
@%1[133]
(37)
The HONO is formed in the reverse of reaction (37) as well
as by reaction (38), which may be homogeneous or heterogeneous, or both (see discussion in references [21] and [138]),
NO
+ NO1 + Hz0 -+
2HONO
(381
and possibly by reaction (39)['341:
H02
+ NO,
+
HONO
+0
2
(391
While HONO has not yet been detected in ambient air, it
has been tentatively identified in smog chamber experiments
on propylene-NO, mixtures in moist air using in situ analysis
by Fourier transform infrared spectroscopy['351.The reactant
concentrations in these experiments were, however, somewhat
higher than ambient (7.8 ppm C3H6,6.2 ppm NO, and 3.8 ppm
Nod.
Angew. Chem. internat. Edit. f Vol. 14 (1975) j No. I
The photolysis of hydrogen p e r ~ x i d d ~ . ~ ' . also produces
OH :
Hi02
+ hv
Ac310nm
w
20H
The first detection and measurement of H202 in ambient
air by a wet chemical method was reported several years
when concentrations as high as 0.18 ppm in Riverside,
California, were measured during a severe smog episode. Subsequent studies on a very smoggy day in July 1973 in Pasadena,
California, located approximately 40 miles west and upwind
of Riverside, using Fourier transform infrared spectrosc0py['~'1, gave inconclusive results on the presence of HzOZ.
Hence further studies on the ambient concentrations and
diurnal variations of H 2 0 2 are important.
The relative importance of each of the OH-forming reactions
changes during the day as discussed in detail by Demerjian,
Kerr, and C a l ~ e r t " ~ ~ ] .
7.2. Kinetics of the Reactions of the OH Radical
Kinetic studies of the reactions of OH with both organics
and inorganics indicate that its reaction rates are very fast,
as shown in Table 3, and in some cases, e.g., reaction with
the higher olefins, are essentially diffusion controlled.
Recently, there has been some controversy over the products
of the important reaction between NO and CH3O2, which
for many years had been presumed to yield NOz and methoxy
indicated
radicals [reaction (42)]. Thus, recent
that the reaction of NO with CH302 occurred by addition
rather than as written in equation (42). However, subsequent
has established reaction (42) as the exclusive reaction
path. The absolute rate of this reaction has not yet been
determined and that of reaction (43)" 491 needs further definition.
7.4. Mechanism of Reaction of OH witb Olefins
The mechanism of the OH-olefin reaction is not known. In
principle either abstraction of a hydrogen atom or addition
to the double bond may occur. However, since activation
energies for abstraction are commonly much higher than those
for addition, one anticipates that at room temperature addition
is the primary mode of reaction, particularly in view of the
very high rate constants (Table 3). This is supported by the
absence of a kinetic isotope effect on the rate of reaction
of C3D6 as compared to C3H6[144.145!In addition, a mass
peak Correspondingto the OH-olefin adduct has been observed
in the reactions of ethylene and propylene at total pressures
of 1-4 torr[1461.These peaks increased with pressure['461sug-
Table 3. Typical rate constants k for some reactions of the OH radical at 300°K.
Compound
Reaction
k[I mol- ' s-
co
0 s
OH+CO-+H+COz
OH + O j - + H 0 2 + 0 2
9.8 x 10'
(3.3 k0.9)x 10'
NO
OH+NOOHONO
( = ~ - Q ) x 109
O H + N 0 2(W
hHN03
OH+CH.++HzO +CH;
OH +HCHO-+H2O+HCO
Ccl
NO2
CH4
HCHO
Ethylene
F'ropylene
trans-2-butene
(C,H3)2C=C(CH3)2
Xylene [d]
[.I
1.1
[CI
[CI
'1
Ref.
7.2 109
6.2 x lo6
( 8 . 4 + 2 . 1 ) 109
~
(1.8+0.6)x lo9
( 8 . 7 k 1 . 3 ) lo9
~
4.5 x 10" [b]
9.9 x 10"' [b]
1 . 1 x lo1"[b]
[a] Most rate studies done at low pressure; this is the recommended value based on extrapolation of rate data
into the second order regime.
[b] Rates measured relative to that for propylene for which k=l.Ox 10'O1mol-'s-l [146].
[c] See discussion.
[dl Mixture of isomers.
7.3. Mechanism of the Reaction of OH with Alkanes and Aldehydes
The reaction of OH with alkanes and aldehydes proceeds
by abstraction to produce HzO and a reactive radical species
such as CH3- in the case of methane, or HCO. on reaction
with formaldehyde. Its rate of reaction with aldehydes is as
fast as that with the smaller olefins12z,144- 1461 and hence
is anticipated to be significant in the atmosphere. The subsequent reactions of the radicals produced may then oxidize
NO to NOz:
(M)
CH, + 0 2 --* CH302
CH302 + N O -+ CH3O + NO2
C H 3 0 + 0 2 -+ HCHO + H 0 2
H 0 2 + N O -+ OH NO2
HCO + 0 2 -t H 0 2 + CO
+
Angew. Chem. internat. Edit. 1 Vol. 14 (1975) f No. 1
gesting that collisional stabilization of the excited adduct is
competitive with its decomposition under these conditions.
The thermochemistry of the addition reaction (44)dictates
that the adduct will have ~ 3 kcal/mol
5
excess energy.
--&-eOH
OH
+ x=C:
l
l
(44)
The corresponding aldehydes are the major stable products,
i.e., CH3CHO and C2H5CHO from the C2H4 and C3H6
reactions,
Isotope studies indicate that the
H atom from the O H is retained in the product aldehyde['461,
suggesting a series of reactions such as (45) and (46). Our
speculations emphasize the need for much more detailed investigations of these reactions under a variety of conditions of
total pressure, oxygen concentration, etc.
Essentially nothing is known about the kinetics and
mechanism of the reactions of OH with arenes, partly because
9
(45)
+-$0. H
+ X
+
+O
C
;{-
HX
H
X = O z etc.
such studies are experimentally difficult. However, this area
is also one which is essential to our understanding of the
detailed course of the complex reaction sequences in polluted
urban air.
8. The Hydroperoxyl (H02) Free Radical
of H 0 2 through a more complex reaction sequence [e.g.
reactions (41)-(43)].
The importance of aldehyde photolysis in photochemical
smog[3,15',
was recently placed on a more quantitative
basis by Culvert and co-workers" 531 who estimated the rates
of H 0 2 generation from HCHO photolysis as a function
of solar zenith angle using the wavelength dependence" s41
of the quantum yields, @I and @II where process I1 [reaction
(47b)] is
HCHO
+ hv
-+
HZ + CO
(11) (47b)
There is presently some controversy about the wavelength
dependence of the absolute and relative values of @I and
@I,['
and clearly, further quantitative photochemical studies
on the system are urgently needed.
8.2. Kinetics of the Reactions of HO, Radicals
8.1. Sources of the HOz Radical
The recognized sources of HOZ in photochemical smog['3sl
include any sources of H or HCO radicals as in aldehyde
photolysis or alkoxy radical oxidations [reaction (43)].
While there have been regrettably few detailed analyses of
the concentrations and types of aldehydes in ambient air,
formaldehyde appears to constitute a significant portion
( z40 %) of the total aldehydes present['501.This will of course
vary with location and nature of the photochemical smog.
For example, Hanst and ~ o - w o r k e r s ~ found
' ~ ' ~ in Pasadena,
California, on July 25, 1973, a peak O 3 concentration of
0.68 ppm and, remarkably, no formaldehyde (50.01 ppm) but
large quantities (0.05ppm) of formic acid.
HOz reacts with a variety of compounds at rates which are
generally several orders of magnitude slower than the corresponding OH reactions. Table 4 gives some typical values
of rate constants recommended by
The errors
quoted for these measurements are in many cases quite large
due to the paucity of reliable data.
Table 4. Recommended [156] values of the rate constants (300°K) for some
gas phase reactions of the HOz radical.
k [lmol-'s-']
Compound Reaction
HOz+CO-+OH+COz
H 0 z + O3+OH + 2 0 2
H 0 z + NO-tOH + NO2
H 0 z +SOz-+OH + SO3
HOz CzHe.-+H20z +CzHs
HOz + HCHO-HzOz + HCO
H02+C2H4-+? [a]
Hoz+i-C4H~-+?[a]
652
~9rf1100%)~105
(2kfactor of 3 ) x 10' [19]
(5.2+ 1.1) x los [19]
(0.08+factor of 10)
(1.7 &factor of 10) x lo3
+
5
5
7 14
[a] See following discussion.
2
- 10
8.3. H0,-Organic Reaction Mechanisms
E
,
"
12
28
6
4
2
$00
2LO
220
260
280
300
h"n1-
320
3LO
360
380
Fig. 4. Absorption spectra [68] for(1) HCHO (25775T); (2) C H K H O (25°C);
and (3) CZHSCHO(25°C) [after [68]).
104
10s
If little is known about the kinetics of HOz reactions, even
less is known about its reaction mechanisms, particularly
with olefins and aromatics. Avramenko and co-workers'' 571
have proposed, from their product studies of the C2H4, C3H6,
and i-C4H8 reactions with H02 in a flowdischarge system
at a total pressure of z 60 torr, that the first step involves addition to the double bond [reaction (49)].
The photochemistry of HCHO is particularly important
because it absorbs radiation well into the actinic ultraviolet
region (Fig. 4) and because the primary process I [reaction
(4741
HCHO
+ hv
-+
H
+ HCO
(1) (47a)
gives two precursors to HOZ radicals per photon absorbed
via reactions (7)and (10). Higher aldehydes are also important
but they yield only one direct precursor of H 0 2 in the primary
photodissociative act I' [reaction (4811
RCHO
+ hv
+
R
+ HCO
(U (48)
and do not absorb as far into the actinic UV as does formaldehyde (Fig. 4). The alkyl radicals may also act as a source
10
(49)
These additions are much less exothermic than the corresponding OH-olefin reactions, with z 10-14 kcal/mol excess energy
residing in the adduct. The isomerization and decomposition
7%
H
CH&-&-H
&OH
-
-
H
CH3-&-CHzO*
bH
(50)
(CH,)~~OH
+ HCHO
Angrw. Chem. intmnat. Edit.
Vol. 14 (1975)
1 No. I
of the adduct [(reaction (50)] and subsequent reactions of
the radical fragments were postulated to give the major
observed products. However, the authors detected 0 atoms
in the flow system and
that the source of the
0 atoms may be the wall reaction
H+Oz %O+OH
Lloyd[’
H+HOz
suggests reaction (52b) may be responsible:
-!%
H2+02
3H20+ 0
% 2OH
In either case, one anticipates that O H will also be produced.
Since both O(3P)and O H react with olefins at rates which
exceed that for H 0 2 by 4-5 orders of magnitude, very small
concentrations of these species may drastically affect the
observed products and product distribution.
It is evident that the entire area of H 0 2 reaction kinetics
and mechanisms warrants extensive investigation. Experiments are most hampered by the problem of producing H 0 2
in a manner which is free from interferences by other reactive
species such as O H ” 561.Wall reactions may also be important
and these are often difficult to factor out. One interesting
approach is that of Hendry and co-workers[1581who have
used the radical initiated gas phase oxidation of 1,4-cyclohexadiene at z 100°C as their source of H 0 2 for kinetic studies.
This method, however, suffers from the disadvantage that
studies must be carried out well above room temperature.
9. Computer and Smog Chamber Simulations of Photochemical Smog
9.1. Computer Kinetic Models
Despite this lack of critical data in many areas, several groups
of researchers have made substantial inroads into computer
simulation of the complex reaction sequences typified by
Figure 1. Ideally, this requires both the detailed reaction
mechanisms and accurate rate constants. For those elementary
reactions whose rate constants are not known, one can estimate
the value by standard methods[1591or else use it as an adjustable parameter which can be varied to give the best fit to
time-concentration profiles (Fig. 1). As pointed out by Calvert
and co-workers[l6’], the latter approach is “a sophisticated
effort in curve fitting”, whereas the former is a more demanding
test of the validity of the proposed mechanism.
It should be recognized, however, that a good computer fit
of the experimental data may be fortuitous and arise from
compensating errors in various parts of the overall scheme,
or from the insensitivity of the total mechanism to that particular elementary reaction.
A detailed computer simulation by Calvert and colleagues[’ 381
illustrates both the utility of this approach and the chemical
insights which can be gained. For example, the predicted
rates of attack (rate=rate constant x concentration of reactive
species) of various reactive species on trans-2-butene in a
“typical” polluted urban atmosphere are given in Table 5.
Using their mechanism and rate constahts, it is evident that
Angew. Chem. internat. Edit.
/ Vol. 14 ( 1 9 7 5 ) 1 No.
1
at short reaction times the rate of attack by O H far exceeds
that by all other species combined and hence O H is primarily
responsible for the excess rate of olefin consumption (Fig.
2) in the early stages of the reaction. At longer reaction times,
0, and HO, become important. It should be stressed, however,
that the validity of these predictions depend on the rate constants and mechanistic details chosen.
Table 5. Rate of attack (ppm min-’) x lo4 on trans-3-butene of some reactive
intermediates as predicted by computer simulation of a sunlight irradiated,
auto-exhaust polluted atmosphere [138]. Initial reactant mixture contains
0.075 ppm NO; 0.025ppm N O z ; 0.lOppm trans-bbutene; IOppm CO:
O.1Oppm C H 2 0 ; 0.06ppm CH3CHO; and 1.5ppm CHI; relative humidity
50% at 25°C; zenith solar angle, 40”.
Time
[min]
WP)
0 3
2
30
0.13
0.17
0.03
0.01
0.26
1.58
I .oo
0.65
90
120
Reactive species
HO2
HO
1.69
1.49
0.49
0.28
18.1
5.9
1.2
0.7
02(’Ag)
2 . 9 10.’
~
2.0x10-’
0.7~10-~
0.4~
lo-‘
9.2. Airshed Models
Research on the detailed kinetic and mechanistic aspects of
photochemical smog such as we have been discussing can
be viewed as simply another “academic exercise” unless the
results can be translated into a form that can be effectively
utilized by public officials charged with the responsibility
of actually controlhg air pollution. Happily this challenge
is currently being met by atmospheric scientists, many of
whom have been cited earlier in this article, who not only
are conducting basic research but also are involved with the
development of reliable and operationally useful airshed
models. These models can be utilized, for example, to assess
the most cost-effective control strategies as well as to predict
well in advance severe smog episodes which may threaten
the health of the public.
Such total airshed models require as input[“‘]: (i) a detailed
emission inventory for fixed and mobile sources; (ii) an assessment of the meteorological and topographical features of fhe
area and an evaluation of the land use; and (iii) a reliable
description of the chemical and physical processes occurring
in the polluted atmosphere. Because of the complexity of
the smog system in the real world, it is generally considered not feasible to include in (iii) all possible reactions
as is done in the so-called “specific mechanism”
approach[’’* 13’* 1621 exemplified above by the work of Caluert
and c o - w o r k e r ~ [ ’ Hence
~ ~ ~ . in airshed models, a “lumped
parameter”[I6’. 1 6 2 ] approach to the chemistry is often used,
i. e., the reaction of a whole class of compounds, e. g. olefins,
with one reactive species, e.g. OH, is represented by one
equation, one rate constant and stoichiometric coefficients
for the amounts of products in each step. In these “lumped
mechanism” models, some of the rate constants and stoichiometric coefficients are then treated as adjustable parameters.
However, this approach suffers from the criticism cited above,
i.e., given enough parameters, curves can be drawn to fit
any set of experimental data. Hence, a model which accurately
reproduces one set of data may not be applicable to ambient
air, or indeed, to another smog chamber.
11
3
9.3. Smog Chamber Studies
A major problem encountered in validating the chemistry
portion of these computer models is the paucity of reliable
data taken over a wide range of carefully controlled experimental conditions in the ppm-ppb concentration range of reactants
typical of polluted ambient air. This gap is understandable
since the technical (and financial) difficulties involved in conducting quantitative mechanistic photochemistry in such
complex systems are not inconsiderable'10.'38.1 6 ' . ' h Z l . Th eY
include, for example, (i) quantitative sampling and analysis
of specific trace contaminants in complex and often heterogeneous mixtures in air, (ii) understanding and dealing with
the important effects of the surface-to-volume ratio and the
nature of the walls of the chamber, (iii) availability of an
ample supply of highly purified air (volumes of some chambers
run into thousands of liters), and (iv) careful control over a wide
range of reaction conditions including temperature, relative
humidity and the intensity and spectral distribution of the
ultraviolet light to which the synthetic smog is actually
exposed.
In order to overcome some of these problems, the chamber
shown in Figures 5 and 6 was constructed'1661.Its features
include (1) FEP Teflon lining to minimize wall reactions,
of
(2) 20mm thick fused silica windows which pass ~ 9 0 %
the 240nm radiation, (3) the capability of evacuation to
torr by a hydrocarbon-free pumping system, (4) a temperature
operation range of -35" to + lOO"C, (5) faithful reproduction
in both absolute intensity and spectral distribution of the
sun's ultraviolet radiation by means of a solar simulator powered by a 25-kilowatt high pressure xenon arc, and (6) an
optical system that generates a collimated light beam that
fills the cylindrical chamber without touching the walls. With
appropriate filters the solar simulator can mimic tropospheric
or stratospheric radiation. Especially important is the match
in intensity in the region 295-330 nm where many important
processes occur, such as the photolyses of ozone, aldehydes,
and hydrogen peroxide. Previous "conventional" chambers
were usually "intensity poor" in this region by virtue of their
light sources (usually fluorescent black lamps emitting primarily 366 nm radiation) and the wavelength transmission cutoff
of the chamber walls.
I
Fig. 6. Schematic diagram of the evacuable smog chamber. I , ambient air
inlet; 2, Teflon coated inner surfaces; 3. pure air inlet; 4, temperature control
manifold (partial); 5, titanium sublimation pump; 6, ion pump; 7, liquid ring
pump; 8, cryogenic sorption pumps; 9, multiple reflection infrared device;
10, light scattering experiment port; 11, sampling ports; 12, quartz windows.
An additional important feature of this chamber is the incorporation of a White cell multiple pass optical system capable
of pathlengths up to 200 meters. When mated to an ultra
high resolution (0.06cm-') and rapid scan (2.5-15pm in
six seconds) Fourier multiplex infrared spectrometer, one can
carry out in-situ analysis of highly reactive pollutants down
to lOppb or less.
2075
I
I
I
2100
2125
2150
A [cm-']
I
I
I
2175
2200
2225
Fig. 7. Infrared spectrum, recorded using Fourier interferometry, of ketene
from the reaction of cis-2-butene (50 ppm) with O 3(10 ppm) at a total pressure
of 2 torr and a pathlength of 50m, in the evacuable smog chambers depicted
in Figs. 5 and 6. The evenly spaced lines are the rotational lines of CO.
For example, Figure 7 shows the high resolution infrared
spectrum of ketene (generated by the reaction of 0 3 with
cis-2-b~tene)['~]
superimposed on the rotational structure of
the 4.7pm band of CO. Ketene has not been detected in
ambient air to date perhaps because of the lack of suitable
analytical methods. If present in smog, however, it may be
amenable to identification by this technique.
10. Concluding Remarks
Fig. 5. Evacuahle smog chamber at the Statewide Air Pollution Research
Center, University of California, Riverside. Length: 3.96 m, diam. 1.52 m,
vol. 55001.
12
We hope that this rather brief and necessarily selective discussion has given the chemist who is not an expert in-the field
some feeling for the problems the air pollution chemist confronts and some of the areas to which he himself can make
a significant contribution. The outcome and utility of air
pollution simulation studies ultimately rest on the accuracy
of the input data generated by a diverse group of scientists
and in the long run the validity of the kinetic models reflects
Angew. Chem. inrtrnar. Edit.
1 Vol. 14 ( 1 9 7 5 ) 1 N o . I
what essentially amounts to the old adage often applied to
computer studies, “garbage in ... garbage out”. It is the task
of we scientists to minimize the “garbage in!”
The authors gratefully acknowledge the generousfinancial assistance ofthe following agencies: California Air Resources Board,
Grant No. ARB-3-01 7and Project Clean Air 122; U.S. Environmental Protection Agency, Grant No. R-800649; Manufacturing
Chemists’ Association, Grant No. 73174-1 ;National Aeronautics
and Space Administration, Ames Research Center, Grant No. NGR
05-008-029; National Science Foundation, Grant No. GP-34524
and GP-38053 X , N S F - R A N N , No. GI-41051, and the Office
of Naval Research, Themis Contract No. N 00014-69-A-02005001. The contents do not necessarily reflect the views and
policies of the above agencies, nor does the mention of tradenames or commercial products constitute endorsement or recommendation for use.- We wish to thank the EOCOM Corporation
and Finniyan Corporation for the loan of a Fourier Multiplex
Spectrometer and a GC-MS respectively, which were used in
portions of the work described above, and our colleagues at
U C R in the Department of Chemistry and the Statewide Air
Pollution Research Center for helpful discussions.
Received: July 9, 1974 [A 32 I€]
German version: Angew. Chem. 87, 18 (1975)
[ l ] J . T Middleton, J . B. Kendrick, and H . W Schalm, Plant. Dis. Rep.
34, 245 ( 1 950).
[2] Third Interim Report on the Smog Problem in Los Angeles County.
Stanford Research Institute, 1950.
[3] P . A. Leighton: Photochemistry of Air Pollution. Academic Press, New
York, N. Y. 1961, and references therein.
141 A. J . Haagen-Smir, E. F . Darley, M . Zaiflin, H . Hull. and W Noble.
Plant. Physiol. 27, 18 (1951).
[S] A . J . Haagen-Smir, Ind. Eng. Chem. 44, 1342 (1952).
[6] A . J . Haagen-Smit, C . E. Bradley, and M. M . Fox, Ind. Eng. Chem.
45. 2086 (1953).
[7] A . J . Haagen-Smif and M . M. Fox, J. Air Pollut. Contr. Ass. 4, 105
( 1 954).
[8] A . P . Alrshuller. S . L. Kopcznski, W A . Lonnemon, 7: L. Becker, and
R. Slatrr, Environ. Sci. Technol. I , 899 (1967).
[9] A. P . Alrshuller, personal communication t o H . Niki, July 1969.
[lo] H . Niki, E . E . &by, and W Weinsrock, Advan. Chem. Ser. 113, 16
(1972). and references therein.
[ 1 I ] F . E. Blacer, Ind. Eng. Chem. 44, 1339 (1952).
[12] R. E . Huir, J . 7: Herron, and D. D . Davis, J. Phys. Chem. 76, 2653
(1972).
[13] M . J . Ball and F . S . Larkin, Nature (Phys. Sci.) 245, 63 (1973). and references therein.
[14] H . Niki, Can. J. Chem. 52, 1397 (1974), and references therein.
[IS] J . Heicklen, K . Westberg, and N . Cohen, Center for Air Environmental
Studies, Report No. 115-69 (1969).
[I61 D. H . Stedman, E. D. Norris, Jr., E. E. Daby, H . Niki. and 8. Weinsrock.
The Role of O H Radicals in Photochemical Smog Reactions. 160th Nat.
Meeting of the Amer. Chem. SOC.,Chicago, IL, Sept. 14-18, 1970.
[I71 D. D. Davis, Can. J . Chem. 52, 1405 (1974).
[I81 W Wong and D. D. Dauis, Int. J . Chem. Kinet. 6, 401 (1974). and
references therein.
[I91 W A. Payne, L. J . Stief, and D. D . Davis, J. Amer. Chem. SOC. 95,
7614 (1973).
1201 G. Dixon-Lewis, W E . Wilson, Jr., and A . A . Westenberg, J. Chem.
Phys. 44. 2877 (1966).
[21] J . G. Caluert: Proceedings of the Conference on Health Effects of
Air Pollutants. National Academy of Sciences, National Research Council,
Oct. 3--5, 1973, pp. 19N.
[22] E. D. Morris, Jr. and H . Niki, J. f h e m . Phys. 5 5 , 1992 (1971).
[23] N . Washida, R. I . Martinez, and K . D. Bayes, Z . Naturforsch. 29a.
251 (1974).
1241 G. H e r z b e r g : Molecular Spectra and Molecular Structures. 111. Electronic Spectra and Electronic Structure of Polyatomic Molecules. Van Nostrand. Princeton 1967, p. 483.
Angrw. Chem. internat. Edit.
J Val. 14 11975) J N o . I
[25] H . W Ford and S . Juffe. J. Chem. Phys. 38, 2935 (1963).
1261 J . N. P i f t s , Jr., J . H . Sharp, and S. I . Chan, J. Chem. Phys. 40. 3655
( 1964).
[27] K . F. Preston and R. J . Cueranouic. Can. J. Chem. 44, 2445 (1966).
[28] I . 7: N . Jones and K . D. Bayes, J. Chem. Phys. 59,4836 (1973).
[29] J N . Pitrs, Jr., 1. H . Sharp. and S . I . Chan, J. Chem. Phys. 39. 238
( 1963).
[30] I . 7: N . Jones and K . D. Bayes. Proc. Roy. SOC.(London) A335. 547
(1973), and references therein.
[31] I . Z N . Jones and K . D. Bayes. Chem. Phys. Lett. 1 1 , 163 (1971).
[32] 1. 7: N . Jones and K. D. Bayrs, J. Chem. Phys. 59, 3119 (1973). and
references therein.
[33] T C. Frankiewicz and R. S. Berry, Environ. Sci. Technol. 6, 365 (1972).
[34] T C. Frankiewicz and R. S. Berry, J. Chem. Phys. 58, 1787 (1973).
[35] E . K . C. Lee and W M. Uselman, Disc. Faraday SOC. 53. 125 (1972).
[36] A . P . AItshulier and J . J. Bufalini, Photochem. Photobiol. 4 , 97 (1965).
1371 D. H . Sfedman, C . H . Wu, and H . Niki, J. Phys. Chem. 77, 251 1 (1973).
[38] H . Niki, personal communication 1974.
1391 D. H . Stedman and H . Niki, Environ. Lett. 4 , 303 (1973).
1401 J . A. Ghormley, R . L. Ellsworrh, and C . J . Hochanadel. J. Phys. Chem.
77, 1341 (1973).
1411 Air Pollution in California. Annual Report 1973. California Air
Resources Board, January 1974.
[42] J . N . P i f t s , Jr., A. C. Lloyd, and J . L. Sprung: Chemical Reactions
in Urban Atmospheres and their Application t o Air Pollution Control Strategies. Proc. Int. Symp. o n Environ. Measurements, Geneva, Switzerland. Oct.
2-4. 1973, and references therein.
[43] L. N . P i f t s , Jr.. A. C . Lloyd, and J . L. Sprung, Chem. in Brit., in
press.
[44] B. J . Finlayson, J . N . Pirrs, Jr., and R. Atkinson, J. Amer. Chem.
SOC.96, 5356 (1974). and references therein.
1451 R. A. Cox and S . A. P e n k e f t , J. C. S . Faraday Trans. 1 68, 1735 (1972).
and references therein.
[46] P . S. Bailey, Chem. Rev. 58, 925 (1958).
[47] R. W Murray, Accounts Chem. Res. I , 313 (1968).
[48] P . S . Bailey, Symposium on Ozone Reactions with Organic Compounds.
161st Meeting of the Amer. Chem. SOC., Div. Petrol. Chem.. Los Angeles.
Calif., March 29-30, 1971; Adv. Chem. Ser. 112, 1 (1972).
[49] W E. Scorr, E. R. Stephens, P, L. H o n s f , and R. C. Doerr, Proc. Amer.
Petrol. Instit. 3, 171 (1957).
[SO] P . L. H a n s f , E. R. Stephens, Ct: E . Scoff,and R . C . Doerr: Atmospheric
Ozone-Olefin Reactions. The Franklin Institute. 1958.
[51] T Vrbaski and R. J . Cuetanouic, Can. J. Chem. 38, 1053, 1063 (1960).
[52] Y K. Wei and R. J . Cuetanouic. Can. J. Chem. 41, 913 (1963).
[53] E . Briner and P. Schnorf, Helv. Chim. Acta l 2 , 154, 181 (1929).
[54] E. Briner and R. Meier, Helv. Chim. Acta 12, 529 (1929).
[SS] J . J . BuJolini and A. P . Affshufier, Can. J . Chem. 43. 2243 (1965).
1561 R. D. Cadle and C . Schadr, J. Amer. Chem. SOC.74, 6002 (1952).
[57] B. J . Finlayson, J . N . P i f f s , Jr., and H . Akimoto, Chem. Phys. Lett.
l 2 , 495 (1972).
[SS] R. Arkinson, B. .; Finlayson, and J . N . P i t f s , Jr., J. Amer. Chem.
SOC.95, 7592 ( I 973).
[59] D. A. Hansen and J . N . P i t f s , Jr.. unpublished results.
[60] W A. Kummer, J . N . Pitls, Jr.. and R. P . Steer, Environ. Sci. Technol.
5, 1045 (1971).
[61] J . N . P i f f s , Jr., B. J . Finlayson, H . Akimoto, W A. Kummer, and R.
P . S f e e r . Proc. Inter. Symp. Ident. Measure. Environ. Pollutants, Ottawa,
Canada, June 14-17, 1971, pp. 32ff.
[62] J . N . Piffs,
J r . , W A . Kummer, R. P . S f e e r , and B. J . Finlayson, Advan.
Chem. Ser. 113, 246 (1972).
[63] S . Toby, J . Luminescence 8, 94 (1973).
1641 H . E . O”ea1 and C . Blumsrein, Int. J. Chem. Kinet. 5 , 397 (1973).
1651 T R. Darling and C . S. Foofe, J. Amer. Chem. SOC.96. 1625 (1974).
and references therein.
1661 R. H i a f t in D. Swern: Organic Peroxides. Wiley Interscience. New
York, N. Y. 1971, pp. Iff.
[67] H. E . O’Neal, personal communication 1973.
[68] J . G. Culvert and J . N . Pirrs, Jr.: Photochemistry. Wiley, New York.
N. Y. 1966, and references therein.
[69] P. G. Blake and G . E. Jackson, J. Chem. SOC.B 1969, 94, and references
therein.
1701 J . M . MrAfee. A . M . Winer, and J . N . Pirts. Jr.: An Infrared Product
Analysis of Ozone-Olefin Reactions by Fourier Interferometry. Paper pre-
13
sented at CODATA Symposium on Chemical Kinetics Data for the Lower
and Upper Atmosphere, Warrenton, VA (USA), Sept. 16-18, t974.
[I151 R. W Murray, J. Lin, and M . L.
1711 E. F. Darfey, E. R. Stephens, J . 7: Middlerun, and P. L. Hansr, Int. J . Air
Pollut. 155 (1959).
[I161 J . W Coomber, D. M . Heberr, W A. Kurnmrr, D. G . Marsh. and
J . N . Pitrs, Jr., Environ. Sci. Technol. 4. 1141 (1970).
[I171 W S. Gleason, A. D. Broadbenr, E . W h i f f l e , and J . N . Pirrs, Jr., J .
Amer. Chem. SOC.92, 2068 (1970).
[72] W N . Arnold, Intern. J . Air Pollut. 2, 167 (1959).
[73] E. A. Schuck, H. W Ford, and E. R . Srephens: Air Pollution Effects
of Irradiated Automobile Exhaust as Related t o Fuel Composition. Rept.
No. 26, Air Pollution Foundation, San Marino, CA (USA) 1958.
[74] E. A. Srhuck and G. J . Doyle: Photooxidation of Hydrocarbons in
Mixtures Containing Oxides of Nitrogen and Sulfur Dioxide. Rept. No.
29, Air Pollution Foundation, San Marino, CA (USA) 1959.
[75] N . A. Renzerri and G . J. Doyfe, 1. Air Pollut. Contr. Ass. 8, 293 (1959).
[76] R. J . Cueranovic, Advan. Photochem. 1 , 115(1963),and references therein.
[77] R. Arkinson and J . N . Pirrs, Jr., J . Phys. Chem. 78,1780(1974).
[78] R. E. Huie, J . 7: Herron, and D. D. Davis, J . Phys. Chem. 76, 3311
(1972).
[79] M. Luria. R. Simonairis, and J . Heicklen, lnt. J . Chem. Kinet. 5, 715
(1973).
[SO] 1. 7: N . Jones and K . D. Bayes, J. Amer. Chem. SOC. 94, 6869 (1972).
[Sl] J . R. Kanofskj and D. Gurmon, Chem. Phys. Lett. 15, 236 (1972).
[82] J . R. Kanofsky, D. Lucas, and D. Gurman, Fourteenth Symp. (Intern.)
on Combustion. The Combustion Institute, 1973, p. 285.
[83] F . J . Pruss, Jr., 1. R. Slagle, and D . Gurman, J. Phys. Chem. 78, 663
( 1974).
[84] S. Saro and R. J. Cueranooic, Can. J . Chem. 37, 953 (1959).
[SS] B. L. Van Duuren, Ann. N. Y . Acad. Sci. 163,633 (1969).
[86] C . Boocock and R. J . Cveranouic, Can. J . Chem. 39, 2436 (1961).
[87] G. R. H. Jones and R. J . Cueranovic, Can. J . Chem. 39, 2444 (1961).
[XX] J. S. Gagney and J. N . Pirrs, Jr.. unpublished data, 1974.
[89] R. A. Bonanno, P. Kim, J . Lee, and R. B. Timmons, J . Chem. Phys.
57, 1377 (1972).
[90] D . Schuerz1e.D. R. Cronn, A. L. Crirrenden, and R. J . Charlson: Molecular
Composition of Secondary Aerosol and Its Origin. Paper presented at 166th
Nat. Meeting of the Amer. Chem. SOC.,Chicago, IL, August 26-31, 1973.
[91] R. J . O'Brien, J . R . Holmes, and A. H . Bockian, Environ. Sci. Technol.,
in press.
[92] E. Grouensrein, Jr., and A . J . Mosher, J. Amer. Chem. SOC. 92, 3810
( 1970).
[93] K . Wei, J . C . Mani, and J . N . Pirts, Jr., J. Amer. Chem. SOC. 89,
4275 (1967).
Koplon, Ann. N. Y. Acad. Sci.
1 7 1 , 121 (1970), and references therein.
[I181 R. A. Ackermon, J . N . Pirrs, Jr., and R. P. Steer, J . Chem. Phys.
52, 1603 (1970).
[I191 R. E. Huie and J . 7: Herron, Int. J . Chem. Kinet. 5, 197 (1973),
and references therein.
[120] J . T Herron and R. E. Huie, Environ. Sci. Technol. 4, 685 (1970).
[I211 1. J . Bufalini, Environ. Sci. Technol. 6, 837 (1972).
[I221 R. F . Firesrone and J . G. Caluerr. t o be published.
[123] C . S. Foore, Science 162, 963 (1968).
[124] C . S. Foore, Accounts Chem. Res. 1. 104 (1968).
[I251 K . Gollnick, Adv. Chem. Ser. 77, 78 (1968).
[I261 K. Gollnick, Advan. Photochem. 6, 1 (1968).
[I271 D. R. Kearns, Chem. Rev. 71. 395 (1971).
[I281 1. R. Polirzer, G. W Griffin, and J . L. Laserer, Chem. Biol. Interactions
3, 73 (1971).
[129] B. Dow,fy,J . L. Laserer, G. W Griffiql. R. Politzer,and C. H. Walkinshaw,
Science 181, 659 (1973).
[130] R. 7:Holman and S. I . Greenberg, J. Amer. Oil Chem. SOC.35,707 (1958).
[I311 R. Corresi and 0.S. Priuerr, Lipids 7, 715 (1972).
[I321 C. C . Wang and L. I. Douis, Jr., Phys. Rev. Lett. 32, 349 (1974).
[I331 H. S. Johnston and R. Graham. Can. J . Chem. 52, 1415 (1974), and
references therein.
11341 R. Simonairis and J . Heicklen, J. Phys. Chem. 78, 653 (1974).
[135] J . M . McAfee, J . N . Pitts, Jr., and A. M . Winer: In-Situ Long Path
lnfrared Spectroscopy of Photochemical Air Pollutants in an Environmental Chamber. Paper presented at the Pacific Conference on Chemistry
and Spectroscopy, San Francisco, CA, Oct. 16-18, 1974.
[I361 J . J . Bufalini, B. W Gay, Jr., and K . L. Brubaker, Environ. Sci. Technol.
6, 816 (1972).
[I371 P. L. Hansr, W E . Wilson, Jr., R . K . Parrerson, B. W Gay, and L.
W Chanej: A Spectroscopic Study of Pasadena Smog. 167th Nat. Meeeting
of the Amer. Chem. SOC.,Los Angeles, CA. March 31-April 5, 1974.
[138] K . L. Demerjian, J. A. Kerr, and J . G. Cafuerr, Advan. Environ. Sci.
Technol. 4, 1 (1974), and references therein.
Nat. Bur. Stand.
[94] C . E. Moore: Atomic Energy Levels. Vol. 1-111,
Circular 4.67, U. S. Government Printing Office, Washington, DC, 1958.
[I391 D. D . Davis, R. SchiK and S. Fischer. unpublished results, as quoted
in ref. [ 171.
[140] J . G . Anderson and F . Kaufman, Chem. Phys. Lett. 19, 483 (1973).
[95] C.-L. Lin and W B. DeMore, J . Photochem. 2, 161 (1973/74), and
references therein.
[96] R. J . Cuetanouic, Can. J. Chem. 52, 1452 (1974).
[141] J. G. Anderson, J . J . Margiran, and F . Kaufman, J. Chem. Phys. 60,
3310 (1974).
[97] R. F. Heidner 111, D. Husain, and J. R. Wiesenfeld, J. C . S. Faraday
Trans. 11 69, 927 (1973).
[98] R. Simonairis and J . Heicklen, Int. J. Chem. Kinet. 5, 231 (1973).
[99] J . N . Pirrs, Jr., R. Arkinson, J. S. Caffney, S. Japar, C. 7: Pore, H .
L. Sandoval, and A. M . Winer, unpublished data.
[IOO] H . L. Sandoval, R. Arkinson, and J . N . Pirrs, Jr., J . Photochem., in
press.
[ I O I ] J . N. Pirrs, Jr., H . L. Sandoual, and R. Arkinson, Chem. Phys. Lett.,
in press.
[I021 M . B. McElroy, S. C. Wol/sy, J. E. Penner, and J . C. McConnell,
J . Atmos. Sci. 31, 287 (1974), and references therein.
[to31 M. J . M o h o and F . S. Rowland, Nature 249, 810 (1974).
[lo41 R. S. Srolarski and R. J . Cicerone, Can. J. Chem. 52, 1610 (1974).
[ 1051 K. D. Bayes, Sixth lnformal Photochemistry Conference. University
of California, Davis, June, 1974, Abstracts, p. 4.
[I061 A. M . Winer and K . D . B a y s , J. Phys. Chem. 70, 302 (1966).
[lo71 C. S. Foore and S. Wexler. J. Amer. Chem. SOC. 86, 3879 (1964).
[IOX] A. U . Khan, J . N . Pirrs, Jr., and E. B. Smith, Environ. Sci. Technol.
I , 656 (1967).
[I091 J . N . Pirrs, Jr., A. U . Khan, E. B. Smirh, and R. P. Wajne. Environ.
Sci. Technol. 3, 241 (1969).
[ I 101 J . N . Pirrs, Jr. in C. S. Tuesday: Chemical Reactions in Urban Atmospheres. American Elsevier, New York, N. Y., 1971. pp. 3ff.
[ I 1 I ] J . N . Pitrs. Jr., Ann. N. Y . Acad. Sci. 171. 239 ( I 970).
[ I 121 R. H. Kummler. M. H . Bortner. and I: Baurer. Environ. Sci. Technol.
3. 248 ( 1 969).
I1 131 R . P. Wayne, Advan. Photochem. 7, 311 (1969).
[ I 141 K. H. We/ge, Can. J . Chem. 52, 1424 (1974), and references therein.
14
[I421 C. Morley and I . W M . Smith, J . C . S. Faraday Trans. I 1 68, 1016
( 1 972).
[I431 W E. Wifson, Jr., J . Phys. Chem. Ref. Data I , 535 (1972).
[I441 F . Stuhl, Ber. Bunsenges. Phys. Chem. 77, 674 (1973).
11451 E. D. Morris, Jr. and H. Niki, J. Phys. Chem. 75, 3640 (1971).
[I461 E . D. Morris, Jr., D. H. Sredman, and H . Niki, J. Amer. Chem. SOC.
93, 3570 (1971).
[I471 C. W Spicer, A. Villa, H. A. Wiebe, and J . Heicklen, J . Amer. Chem.
SOC.95, 13 (1973).
114x1 C. 7: Pare, B. J. Finlayson. and J . N. Pirrs, Jr., J . Amer. Chem.
SOC.96, 6554 (1974), and references therein.
[I491 J . Heicklen, Advan. Chem. Ser. 76, 23 (1968).
[I501 A. P. Alrshuller and S. P. McPherson, J . Air. Pollut. Contr. Ass. 13,
109 (1963).
[I511 A. P. Alrshuller, 1. R. Cohen, and 7: C . Purcell, Science 156, 937
( 1967).
[152] J . N . Pirrs, Jr., 1. Air Pollut. Contr. Ass. 19, 658 (1969).
[153] J . G. Caluerr, J: A. Kerr. K . L. Demerjian, and R. D. McQuigg, Science
175, 75 1 ( 1 972).
[154] R. D. McQuigg and J . G . Caloerl. J. Amer. Chem. SOC. 91, 1590
( 1969).
[l55] H. P. Sperling and S. Toby, Can. J . Chem. 51,471 (1973), and references
therein.
[156] A. C. Lloyd, Int. J. Chem. Kinet. 6, 169 (1974).
[157] L. I . Auramenko, L. M. Eulashkina, and R. V Kolesnikova, Bull. Acad.
Sci. USSR, Div. Chem. Sci. (Transl. of Izv. Akad. Nauk SSSR. Ser. Khim.)
1965, 1306.
[158] D. G. Hendry and W R. Mabey: A New System for Determining
Rate Constants for H 0 2 Reactions. 167th Nat. Meeting of the Amer. Chem.
SOC.,Los Angeles, CA. March 3!-April
5, 1974.
Angew. Chem. internat. Edit. Val. 14 (1975)
1 No.
1
11591 S. W. Benson: Thermochemical Kinetics. Wiley, New York, N. Y.,
1968.
[I601 J . G. Caluert, K . L. Demerjian, and J . A . K e r r , Proc. Int. Symp.
Air Pollut., Tokyo, Oct. 17-19, 1972, pp. 465ff.
[161] A. Q . Eschenroeder and J . R . M a r f i n e z , Advan. Chem. Ser. 113, 101
(1972).
[I621 7: A . Hecht, J . H . Seinfeld. and M. C . Dodge, Environ. Sci. Technol.
8, 327 (1974). and references therein.
[I631 C. M. Shy and J . F . Finklea, Environ. Sci. Technol. 7, 205 (1973).
[I641 R . P. Steer, K . R . Darnall, and J . N . Pifrs, Jr., Tetrahedron Lett.
1969. 3765.
11651 S. Furuyama. R . Atkinson, A . J . Colussi, and R . J . Cvetanouic, Int.
J. Chem. Kinet. 6, 741 (1974).
[166] J . N . Pitrs, Jr., P. G. Bekowies. G. J . Doyle, J . M . McAfee, and
A. M. Winer, to be oublished.
Umpolung of Amine Reactivity. Nucleophilic a-(Secondary Amino)alky lation via Metalated Nitrosaminesr**lr”’l
New synthetic
methods 0
By Dieter Seebach and Dieter E n d e d * ]
There are basically two kinds of hetero atoms in organic molecules: one kind confers electrophilic
character upon the carbon atom to which it is bound, and the other kind turns it into
a nucleophilic site. The development of methods permitting transitions between the two resulting
categories of reagents has become an important task of modern organic synthesis. The scope
of such umpolung of the reactivity of functional groups is discussed for the case of amines
as an example. A method of preparing masked u-secondary amino carbanions consists in
nitrosation of the secondary amine, followed by metalation of the resulting nitrosamine CL
to the nitrogen, reaction with electrophiles, and subsequent denitrosation. Many examples
are given for each of these steps which illustrate the wide scope of the overall synthetic
operation (electrophilic substitution at the a-C atom of the secondary amine). Preliminary
applications and a method for avoiding the handling of nitrosamines are presented, and the
report concludes with a brief account of the significance of nitrosamines in the study of
carcinogenesis and mutagenesis.
reagents possessing a given “philicity”. Nucleophiles attack
the odd C atoms (N’. 3 , . . . attack)[’’whereas electrophiles attack
the even C atoms ( E 2 * 4 , . .attack)
.
of the chain (for examples
see Table 1).
1. Introduction
1.1. Concept of Umpolung
We chemists are obliged to conform to certain simple rules
dictated by Nature. Thus from the very beginning of our
education in organic chemistry we learn-just how explicitly
depends on the quality of our teacher-that
heteroatoms
such as halogen, oxygen, and nitrogen produce a certain reactivity pattern in a carbon skeleton : their electronegativity, being
greater than that of carbon (-I effect), and their ability to
stabilize an adjacent positive charge ( +M effect) lead to the
sites shown in Scheme 1 which are susceptible to attack by
In C-C linking reactions of two or more heterosubstituted
carbon compounds it is thus established what synthetic opera-
+
sites of electrophilic
nucleophIIIcreactivity are C atoms
attack by nucleophiles
electrophilesoccurs at C atoms
0
OH
‘*
3, 5, ...
2, 4, 6, ._.
Scheme 1. Polarity pattern in hetero-substituted carbon chains.
[‘I
Prof. Dr. D. Seebach and Dr. D. Enders [**I
Institut f i r Organische Chemie der Universitat
63 Giessen, Ludwigstrasse 2 I (Germany)
[**I This report draws extensively on the Dissertation of D. E.
[**‘I Attempts to find an English equivalent of the term “Umpolung” have
yielded none so concise as the original German word. Its general adoption
into English has therefore been proposed (see Chem. Ind. (London) 1974,
910).
Angew. Chem. internat. Edit.
1 Vol. 14 ( 1 9 7 5 ) 1 NO. 1
[*I Wherever there is a danger ofconfusion with a nitrogen atom the nucleophile will be designated “Nu”.
15
Документ
Категория
Без категории
Просмотров
0
Размер файла
1 495 Кб
Теги
air, mechanism, pollution, photochemical
1/--страниц
Пожаловаться на содержимое документа