ethyl nitrite. The concentrations of naphthalene and ethyl
nitrite were determined in control experiments without
photolysis for fixed temperatures of the naphthalene and ethyl
nitrite reservoirs and fixed gas flow rates, as reported
previously (20). A 4.6-W low-pressure mercury arc lamp was
used for the photolysis of ethyl nitrite. Reaction products
were collected in a cold trap on the exit side of the reactor.
LC/MS. A Hewlett-Packard Series II Model 1090 liquid
chromatograph was interfaced to a VG Quattro II mass
selective detector operating with MassLynx software. The
mobile phase was a gradient of methanol and acetic acid/
ammonium acetate buffer, pH 3.6, and was used for both
positive and negative ion electrospray LC/ MS.
Reagents. HPLC grade methanol, dichloromethane, isooc-
tane, spectranalyzed grade THF, ACS reagent grade anhydrous
ethyl ether, 1,4-dioxane, tetrahydrofuran, isopropyl alcohol,
acetic acid (99.8%), and sulfuric acid (Fisher) were used as
received, except that THF was purified by keeping it overnight
over potassium hydroxide pellets, followed by distillation in
a flame-dried apparatus over sodium metal under nitrogen.
Authentic standards and reagents for organic synthesis were
obtained from Aldrich. 1,4-Naphthoquinone, benzoic acid,
phthalide, phthalaldehyde, and 2,4-dinitro-1-naphthol were
purified using sublimation and had purities >99% by GC/ MS
or HPLC. 2-Nitronaphthalene was recrystallized from etha-
nol, mp, 72-74 °C, lit. (22) 79 °C, purity >99.9% by GC/ MS.
Syntheses. Isopropyl Nitrite (23, 24). To a mixture of 30
mL of water and 40 mL (0.75 mol) of concentrated sulfuric
acid in a 250-mL Erlemeyer flask, cooled to 0 °C, was added
Sm og Cham ber. The smog chamber was a cylindrical,
3
9
2
.4 m , flow-mode design, constructed from Teflon film with
m diameter Teflon-coated aluminum end plates. A full
description of the chamber has been published elsewhere
21). The lamps were a combination of Phillips fluorescent
UV-A and Sylvania blacklights, having λmax ∼ 375 nm and
J(NO at λ > 350 nm with all the lamps on.
) ) 1.2 × 10
(
-
3
-1
2
s
This system was designed to optimize the production of
hydroxyl radicals rather than to simulate sunlight. Ports were
fitted on the end plates of the chamber to allow the entry and
exit of gases, and the contents of the chamber were mixed
by means of an interior fan. The chamber was flushed with
pure dry air for 2 days prior to each experiment, with the
lamps on at maximum intensity, in order to avoid contami-
nation from previous experiments. An on-line Hewlett-
Packard Model 5890 Series II gas chromatograph with a
packed 8 ft × 1/ 8 in stainless steel packed column (OV17 on
Chromosorb W) and flame ionization detector was used to
quantitate the initial and unreacted amounts of reactant
during kinetic runs. Samples were introduced into the GC
by means of a 5-mL sampling loop.
1
15 mL (1.5 mol) of ACS grade (97%) isopropyl alcohol. The
solution was allowed to warm to room temperature, trans-
ferred to an equalizing-pressure funnel with ground glass
stopper, added dropwise over 2 h into a solution of sodium
nitrite (114 g, 1.65 mol) in 500 mL of water in a three-neck
flask, and cooled to -5 °C in an ice/ salt bath. The mixture
was stirred in the cold for a further 1 h, after which time the
Off-line Analysis. Samples to be analyzed at the University
of Guelph were condensed out of the chamber using a
proportional flow meter and liquid nitrogen-cooled impinger;
by recording the flow rate and collection time, the concen-
tration of material in the cold trap could be related back to
the concentration in the chamber. The contents of the trap
upper (yellow) layer was separated, dried over Na
2 4
SO , and
distilled (bp 39-40 °C). The product was kept, foil-wrapped,
in the refrigerator. The synthesis of ethyl nitrite was similar,
except that the product (bp 16-17 °C) was simply separated
from the reaction mixture and used without distillation.
2
,3-Epoxy-2,3-dihydro-1,4-naphthoquinone (25). Ten mil-
2 2
were rinsed out using CH Cl (for GC/ MS analysis) or
liliters of 5.25% sodium hypochlorite (NaOCl) was added to
a solution of 1.0 g of 1,4-naphthoquinone in 20 mL of 1,4-
dioxane in a water bath at 0 °C. After 3 min, the reaction was
quenched by adding 35 mL of water. The crude product was
filtered and recrystallized from ethanol, mp 132-133 °C (lit.
methanol (for HPLC analysis). The solvent (7-9 mL) was
blown down with ultrapure nitrogen before analysis, and the
residue was made up to 100 µL in a V-shaped vial before
analysis. GC/ MS was carried out using a Hewlett-Packard
Series II GC equipped with 30 m J&W DB-5MS column with
film thickness of 0.25 µm, interfaced to a Hewlett-Packard
Model 5971 mass selective detector operating in the EI mode.
The internal standard was 1-naphthylbutyrate, which was
added in weighed amount to a weighed quantity of the cold-
trapped product mixture. The standard was chosen to elute
late in each run so as not to interfere with product peaks.
Later, a Hewlett-Packard Model 5972 mass selective detector
was used to allow off-line GC/ MS, and the column was
replaced by a capillary 25 m × 0.2 mm column coated with
(
25) 134-136 °C), 99.9% pure by GC/ MS; m/z (%): 74 (29),
+
7
1
2
6 (54), 77 (34), 89 (60), 105 (100), 146 (38), 173 (51), 174 (M ,
1
); H NMR (CDCl
H), 7.762 (dd, J ) 6.0, 3.2 Hz, 2H), 4.012 (s, 2H).
E)-2-Formylcinnamaldehyde. 2-Chlorocarbonyl-trans-
cinnamyl chloride was prepared from 2-carboxy-trans-
cinnamic acid (2 g, 10 mmol) and PCl (4.5 g, 21 mmol) as
3
) 400 MHz: δ 7.895 (dd, J ) 6.0, 3.2 Hz,
(
5
reported by Elvidge and Jones (26). The mixture was heated
at 100 °C until evolution of hydrogen chloride ceased, and 10
mL of petroleum ether (bp 80-100 °C) was added. The acid
chloride crystallized over night and was dried in vacuo to
give 0.93 g (4.5 mmol, 45%) of needle-shaped crystals that
5
µm of phenylmethylsilicone oil (5% phenyl). The following
were typical temperature programs: DB-5MS column, 100
-
1
-1
°
C (11.0 min), then 3 °C min to 170 °C, then 10 °C min
were 97% pure, as determined by GC/ MS; mp 63-65 °C (lit.
-
1
to 250 °C, then 20 °C min to 290 °C; phenylmethylsilicone
column, 100 °C for 3 min, then 5 °C min to 250 °C, then hold
for 5 min at 250 °C.
+
(
(
27) 70 °C); GC/ MS mass spectrum m/z (%): 228 (M ), 195
-
1
33), 193 (98), 167 (34), 165 (100), 157 (37), 130 (24), 129 (26),
1
02 (50), 101 (41), 75 (25).
HPLC was carried out using a Gilson HPLC equipped with
binary pumps, Model 805 manometric module, Model 811B
dynamic mixer, Waters Model 441 UV absorbance detector
The crude 2-chlorocarbonyl-trans-cinnamyl chloride was
reduced to the aldehyde by the method of Herbert and Subba
Rao (28). The acid chloride (0.93 g, 4.5 mmol) was dissolved
(
254 nm), Waters Model 710 injector, and Varian Model 4290
in 5 mL of dried THF and placed in a flame-dried, N -flushed
2
integrator. The column was a Waters µBondpak reverse phase
25-mL flask fitted with a stirrer and 10-mL syringe, and cooled
to -78 °C. Lithium tri-tert-butoxyaluminohydride (9 mmol
in 9 mL of dried THF) was added over a period of 15 min,
avoiding any major rise in temperature. The mixture was
allowed to warm to room temperature, and ice water (4 mL)
was injected into the flask. After several extractions with
ether and removal of the ether in vacuo, the product was
purified by TLC (silica gel, using 1.5:1, ether:isooctane) to
C
18 of dimensions 3.9 mm × 300 mm. Solvents were HPLC
grade, filtered through a 0.45-µm Nylon 66 membrane and
degassed before use. Retention times were recorded relative
to that of cinnamic acid. Early experiments were carried out
using mixtures of water and methanol; the results in Table
2
were obtained using methanol (solvent A) and CH
3 2
CO H/
CH CO NH buffer, pH 3.6 (solvent B) using the following
3
2
4
program: 25% A with linear gradient to 1:1 A:B over 6 min;
hold at 1:1 A:B for 10 min, then linear gradient to 100% A over
give 22 mg (14%) of 2-formylcinnamaldehyde (29), R
mp 57-58 °C; GC/ MS m/z: 77 (23%), 103 (31), 131 (100), 160
f
) 0.46,
+
1
1
4 min.
(M , <1); H NMR (CDCl ) 200 MHz: δ 6.52 (dd, J ) 8, 16 Hz,
3
VOL. 31, NO. 8, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2 2 5 3