:It 13262
Proceedings of Healthy Buildings 2000, Vol. 1
567
AIR QUALITY MEASUREMENTS IN A MODEL KITCHEN USING
GAS AND ELECTRIC STOVES
2
Barbara Fluckiger1, Martin Seifert , Theo Koller1, Christian Monn 1
1 Institute for Hygiene and Applied Physiology, Swiss Federal Institute of Technology, Zurich
2
Swiss Gas and Water Industry Association, Zurich
ABSTRACT
Epidemiological studies suggest that cooking with gas leads to an increase of air pollutants
and may enhance symptoms of respiratory diseases. However, little experimental data are
available concerning the emission of pollutants due to different cooking processes. The goal
of this study was to investigate the influence of cooking under standardized conditions on the
indoor air quality. A model kitchen was built and three different gas stoves and one electric
stove were included in the study. Two different menus were prepared with different settings
of the ventilation rate of the exhaust. Online measurements of NO, N02, CO, particles,
polyaromatic hydrocarbons and climatic factors were performed on the side of the stoves. The
results showed large differences between different gas stoves with respect to emission of
gaseous pollutants. Particle concentrations depended strongly on the menu cooked
independent of the stove used. An adequate exhaust hood and ventilation rate are required in
order to reduce the air pollutants, in particular the fine particles, to tolerable concentrations.
KEYWORDS: Gas appliances, kitchen exhaust hood, carbon monoxide, nitrogen oxides,
particulate matter, ventilation
INTRODUCTION
In Switzerland, out of
2,5
million households some
20% are supplied with natural gas,
mainly
for cooking. Epidemiological studies suggested that cooking with gas leads to an increase of
air pollutants (mainly NO, N02 and CO) and may enhance symptoms of lung diseases.
However, in the numerous studies conducted during the past both positive and negative
associations between symptoms and pollutants levels have been reported
[1].
Most positive
studies indicate, that effects are small and difficult to detect. Many studies concentrated on
children which are believed to be more susceptible to effects of indoor air pollution than
adults
[2].
Since recent studies showed that peak concentrations may be more important for
health effects than long-term concentration averages
[2] the persons doing the cooking at the
stoves are primarily exposed and at risk. This was confirmed in a study where associations
were found between the use of gas stoves and lung function decrements and increased
respiratory symptoms for young women but not for men living in the same homes
(3).
Different (mainly epidemiological) studies reported higher N02 and particle concentrations in
kitchens and living rooms of homes with gas-cooking compared to the homes with electrical
stoves
[4-7). Studies with focus on the kind of stoves used showed that pilot lights are a main
contributor for increased N02-concentrations and that large differences could be found
between kitchens with gas ranges and ovens and those with gas ranges only
[8, 9).
568
Exposure, human responses and buiding investigations
However, data about the amount of pollutants emitted during cooking of different foods under
different ventilation conditions are missing. The goal of this study was to investigate the
influence of cooking under well standardized conditions on the air pollution concentrations
inside a model kitchen.
METHODS
For the experiments, a model kitchen was designed and built in the laboratory. The kitchen
was of a wooden construction
(3 x 4 m, 2.3 m high) and equipped with standard appliances.
Three gas stoves of different design and one electric stove were included in the study. The
different stoves were installed underneath a kitchen exhaust hood with variable settings of the
3
ventilation rate (90, 155, 200 and 260 m /h). The main experiment consisted in the
preparation of two different common menus (chicken curry with rice or a potato pancake
(Rosti) with fried egg) with each of the stoves and with the four different ventilation levels in
operation. All the experiments were conducted in duplicates.
The samples for the measurements were taken nose high on the right side of the stove. The
investigation included online gas measurements for NO, N02 and CO (Monitor Labs, APMA300E), different particle measurements (Grimm Dust Monitor for counts per liter; PM10) and
polyaromatic hydrocarbons (PAH) measurements, climatic factors (temperature, relative
humidity, wind velocity) and passive samplers for N02 and formaldehyde. The consumption
of gas and electricity were also measured concomitantly.
RESULTS
Measuredpollutants and influence of the different menus
The menus "Curry" and "Rosti" were cooked with standardized procedures on the three gas
stoves and on the electrical stove with each of the four different ventilation rates in operation
twice. Cooking of the menus on the gas stoves increased the concentrations of CO, NO and
N02 (Figure ia). The amount emitted pollutants depended on the amount of gas burned
during each experiment, the number of burners in use and on the type of stove.
During all the cooking experiments, i.e. also upon using the electric stove, particle counts
increased (Figure lb). Especially in the menu Curry, when meat was fried in the pan the
particles reached extremely high concentrations,
10-20 times higher than with the Rosti menu.
During the frying of the potato chunks (Rosti) in the pan the temperatures were lower than in
the Curry menu and probably less oil or water aerosols were generated.
The PAR-concentrations were also dependent on the menu cooked. They were mainly
elevated during frying meat. In the experiments with the Curry menu an increase up to 50
3
ng/m was seen in one case with ventilation rate 1 (Figure la), but usually only slight
3
increases occurred (average increase during the whole cooking period of about 5 ng/m ).
Hardly any change in PAR-concentration was observed during the preparation of the Rosti
menu and in most cases where the oven was used.
Temperature and relative humidity inside the kitchen raised depending on the amount of gas
burned (different for the two menus). Rice was cooked with the Curry menu and the steam of
the boiling water increased the relative humidity by a few percents depending on the actual
ventilation rate setting (data not shown). At the end of the cooking all concentrations dropped
rapidly back to baseline concentrations while the exhaust was still in operation (Figure
1).
Proceedings of Healthy Buildings 2000, Vol. 1
569
a.
Concentrations during cooki ng of the Curry menu, ventllatlon rate 1
300
4500
.........................................................................................................................................................
4000
250
3500
200
J:
�
3000
co
"'
�
2500 0
0
.a
c
8
150
z
0
z
.a
cooking start
8:
\
end of cooking
·
2000
CL
CL
1500
100
1000
50
500
0
Ill
r--
0
� � '!l
� �
�
"' C\J Ill
v <D "'
'!l 0 0 � � � � �
IN IN IN IN IN IN IN
::
�
�
�
�
�
�
Ill
�
C\J
C\J
�
�
�
Q)
C\J
r--
0
;;; � � �
IN IN C\J C\J
IN �
�
�
�
�
� � � iA ;1;
C\J C\J C\J C\J IN
�
�
�
r-Ill
IN
�
0
�
..,
�
b.
Particle counts during cooking of the Curry menu, ventllation rate 1
250000
,
..........•.....•......................•.•..............•.................................................•...•..........................................
200000
particles> 0.75 µm
·ii
l5
.. 150000
cu
=
.'!
c
"
0
"
.!!
"
'€
"'
particles > 1 µm
100000
CL
50000
0
"' r-� -,;
:?l � -�
Figure
"'
If!
-
Q)
� � 0
� !!! �
v
co
� � �
"'
i:j
r-� � � (;; � "'
� '::J
� � � � c-i c-i
c-i c-i
Ill
v
r--
� u; ;1j Ill
c-i c-i c-i � c-i
8
�
1. Concentrations of CO, NO, N02 and PAH (a) and particles (b) during cooking of
the Curry menu on gas stove 1 and with ventilation rate 1 in operation. A total of 80 liters of
gas was burned.
570
Exposure, human responses and buiding investigations
Differences between the stoves
Experiments without and with a pan but no cooking except water on top of the burners and
the experiments with the cooking of the two menus showed that large differences between the
four different stoves and the measured concentrations of pollutants occurred (Figure 2). While
only the cooking of the menus with gas stoves leads to increased concentrations of NO, N02
and CO in the kitchen air, the highest particle concentrations were measured during cooking
on the electrical stove. For comparison of the four stoves used in the experiment of Figure 2,
the same menu was cooked to the same extent as closely as possible.
Curry: Maximal particle counts reached
Curry: Maximal CO-, NO-& N02-
(ventllatlon rate 3)
160000 .................................................. . .......... !
3)
120.0
concentrations reached ( ventllation rate
2000
: ..··-··························....................
·-·-·····-·· -·
oco
100.0 llN02
Iii NO
1000
1600
1400
;
BO.O
1200
E
= 1000
:E
0
(.)
.a
c.
c.
c-----+ 60.0
BOO
600
t-----!
400
40.0
20.0
200
0.0
o
Gas 1
Gas 2
Gas 3
Electric
Stove
;
E
;<
..
:E
..
0
z
oil
0
z
.a
c.
c.
. .
140000
+------__...,��!
-; 120000 +-----�
..
� 100000
!!
--------­
Size in µm
[j>10
02 to 10
111 to 2
firil>0.75to1
c
i: 80000
<>
..
�
..
...
+------
60000
40000
20000
0
Gas
1
Gas
2
Gas 3
Electric
Stove
Figure 2. Maximal concentrations of NO, N02 and CO as well as particle counts during the
cooking of the Curry menu with the three different gas stoves and the electrical stove.
(Baseline concentrations: 130-280 ppb CO; 9-11 ppb NO; 5-8 ppb N02 and 500-4000
particles> 0.75 • m). The same Curry menu was cooked on the 4 different stoves under
standardized conditions.
Influence of the use of a gas oven
If no kitchen exhaust was in operation the concentrations of CO, NO and N02 increased
steadily during the use of a gas oven. In the experiments with an oven temperature of 200°C
during 50 minutes the maximal concentrations were 2-20 ppm CO, 750-2000 ppb NO and
350-700 ppb N02, depending on the type of oven in use. With ventilation rate 1 in operation
the concentrations only reached 20-30 % of the maximal concentrations without ventilation.
With the electrical oven there was no increase in NO- and N02-concentrations, but CO­
concentrations slightly increased. Except for a raise in temperature and relative humidity no
significant effects on particle concentrations or other pollutants were seen during the
experiments with the gas or electrical ovens.
Effect of ventilation
Figure 3 shows the effect of the ventilation rate setting of the exhaust hood on the pollutant
3
concentrations. Setting 2 (155 m /h) reduces these concentrations significantly compared to
3
setting 1 (90 m /h). A further increase in the ventilation rate (settings 3 and 4) does not
necessarily lead to a further improvement of the air quality.
Proceedings ofHealthy Buildings 2000, Vol. 1
Curry menu: Increas e In CO-, NO and N02-
Ros ti menu: Increa s e In CO-, NO and N02-
concentratlon for d ifferent ventllatlon rates
concentratlon for d ifferent ventllatlon rates
2500
.............
,,...., ......
..........................,........
.,,,
.
...... .,.....
__.. ...........
. .,
, ..
2000
0
0
.ti
Q.
Q.
571
1500
1000
500
180
160
140
120
100
80
60
40
20
2500
2000
•N02
N
0
z
mNo
-g
c
"
0
z
0
0
.ti
Q.
Q.
.ti
Q.
Q.
V1
V2
V3
1500
1000
0
V4
N
0
z
BN02
�NO
-g
c
"
0
z
.ti
Q.
Q.
V2
V3
V4
Ventilation rate
Rostl menu: Increase In partlcle counts
Curry menu: Increas e In particle counts
Ventllatlon rata
oco
0
V1
Ventllstlon rate
100000 -..----.... ,...,...
... ..,,. ..,...
... --,,
Size in µm
90000
..
11> 10
80000
"
..
02 to 10
GI
70000
:!:::
m110 2
c::
60000
Ill
....
1.!11> 0.75 to 1
c
50000
"
0
u
40000
.!!
u
30000
:e
"
20000
Q.
10000
0
V4
V3
V2
V1
180
160
140
120
100
80
60
40
20
500
0
0
o0010I0000000000000000000HOO OU OO l00100000001 0>000
IJCO
10000
--,
�----- ,...
..,......,,
....,Size in µm
..
[i]> 10
"
..
GI
.t:
c::
Ill
...
c
"
0
u
02 to 10
111 to 2
if> 0.75 to 1
.!!
u
�
..
Q.
V1
V2
V3
V4
Ventllatlon rate
Figure 3. Average increase in CO-, NO and N02-concentrations (ppb) and average increase in
particle counts per liter of air during cooking of the two menus Curry and Rosti with gas stove
1 under different ventilation settings (Vl to V4 ) .
The location of the gas burner used for the cooking was important for the ventilation
efficiency. Using a gas burner in the front caused a higher increase in pollution concentrations
than a burner in the back. The evaluation of the ventilation flow showed that the exhaust was
much more effective in the back half (75% of the total air flow) than in the front row of the
stove.
Since in many kitchens the exhaust ventilator is integrated in a row of cupboards on the wall
experiments were conducted to simulate this situation. Two pieces of cardboard (30 cm high)
were tightly attached along the side of the kitchen exhaust hood. The experiments with the
menus showed that this modification increased the efficiency of pollutant removal
enormously and reduced the pollutant concentrations by about 50 % under the same
ventilation rate setting (data not shown).
572
Exposure, human responses and buiding investigations
DISCUSSION
Our results showed very short concentration peaks for NO, N02 and CO during the cooking
of a menu, but with large differences using different gas stoves. Hourly means of these
emissions were much below WHO guidelines [10]. However, particle concentrations reached
during the cooking were high and depended strongly on the menu cooked and on the
ventilation performance. They tended to be higher with the electric than with the gas stoves,
probably because of the slower temperature adjustments with the electrical heating than using
a gas burner.
The design of the kitchen exhaust proved to be poor and only slight improvements of the
pollution situation in the kitchen were reached with increased ventilation rates. Simple
improvements of the design of the exhaust hood (small covers on the side) allowed to reduce
pollutant concentrations in the ambient air during cooking to tolerable levels.
The main contribution of NO, NOi and CO comes from the use of gas ovens. Here it is
important that adequate ventilation is provided (better instructions of the users and/or
technical measures).
ACKNOWLEDGEMENTS
The study was funded by the Swiss Gas and Water Industry Association (SGWIA). The
authors would like to thank the staff of the SGWIA-group in Schwerzenbach for their support
during the installation of the model kitchen and the measurements.
REFERENCES
347: pp 412.
l.
Brauer, M. and Kennedy, S.M. 1996. Gas stoves and respiratory health. Lancet.
2.
Garrett, M.H., Hooper, M.A., Hooper, B.M., et al. 1998. Respiratory symptoms in children
and indoor exposure to nitrogen dioxide and gas stoves. Am J Respir Crit Care Med. 158(3):
pp 891-895.
3.
4.
Jarvis, D., Chinn, S., Luczynska, C., et al. 1996. Association of respiratory symptoms and lung
function in young adults with use of domestic gas appliances. Lancet. 347: pp 426-431.
Melia, R.J.W., Chinn, S., and Rona, R.J. 1989. Indoor levels of N02 associated with gas
cookers and kerosene heaters in inner city areas of England. Atmospheric Environment.
5.
24b(l):
pp 177-180.
Coward, S.K.D. and Raw, G.J. 1995. Nitrogen Dioxide Levels in Homes in Avon. in Proc. of
Healthy Building '95.
6.
Lambert, W.E., Samet, J.M., Hunt, W.C., et al. 1993. Nitrogen dioxide and respiratory illness
in children: Part II: Assessment of exposure to nitrogen dioxide. Res Rep Health Eff Inst. 58:
pp 33-50.
7.
Braun-Fahrlander, C. 1989. Auswirkungen von Luftschadstoffen au/ die Atemwege von
Kleinkindern. Schweizerische Medizinische Wochenschrift. 119: pp 1424-1433.
8.
Spengler, J., Schwab, M., Ryan, P.B., et al. 1994. Personal Exposure to nitrogen dioxide in
the Los Angeles Basin. J. Air Waste Manage. Assoc.
44(1) : pp 39-47.
9.
Raw, G.J. and Coward, S.K.D. 1995. Factors affecting nitrogen dioxide levels in homes. in
10.
Proc. bf Healthy Buildings '95.
WHO. 1995. Update and Revision of the Air Quality Guidelines for Europe, . World Health
Organization, Regional Office Europe: Copenhagen, Denmark.