CC BY-NC-ND
0Accepted Manuscript
Emissions of Hydrogen Cyanide from On-road Gasoline and Diesel Vehicles
Samar G. Moussa, Amy Leithead, Shao-Meng Li, Tak W. Chan, Jeremy J.B. Wentzell, Craig Stroud, Junhua Zhang, Patrick Lee, Gang Lu, Jeffery R. Brook, Katherine Hayden, Julie Narayan, John Liggio
PII: S1352-2310(16)30084-X
DOI: 10.1016/j.atmosenv.2016.01.050
Reference: AEA 14430
To appear in: Atmospheric Environment
Received Date: 8 July 2015 Revised Date: 26 January 2016 Accepted Date: 27 January 2016
Please cite this article as: Moussa, S.G., Leithead, A., Li, S.-M., Chan, T.W., Wentzell, J.J.B., Stroud, C., Zhang, J., Lee, P., Lu, G., Brook, J.R., Hayden, K., Narayan, J., Liggio, J., Emissions of Hydrogen Cyanide from On-road Gasoline and Diesel Vehicles, Atmospheric Environment (2016), doi: 10.1016/ j.atmosenv.2016.01.050.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
HCN Emission
HCN Detection
Laboratory and ambient measurements of hydrogen cyanide (HCN) from vehicles using high resolution Proton Transfer Reaction- Time of Flight- Mass Spectrometry (PTR-TOF-MS)
1 Emissions of Hydrogen Cyanide from On-road
2 Gasoline and Diesel Vehicles
5 Samar G. Moussa, *1 Amy Leithead,1 Shao-Meng Li,1 Tak W. Chan,2 Jeremy J.B. Wentzell,1
6 Craig Stroud,1 Junhua Zhang,1 Patrick Lee,1 Gang Lu,1 Jeffery R. Brook,1 Katherine Hayden,1
7 Julie Narayan,1 and John Liggio *1
9 Atmospheric Science and Technology Directorate Science and Technology Branch,
10 Environment Canada, 4905 Dufferin Street, Toronto, Ontario, Canada M3H 5T4
11 Emission Research and Measurement Section, Environment Canada, 335 River Road, Ottawa,
12 Ontario, Canada, K1A 0H3
13 *Author to whom correspondence should be addressed: John.Liggio@canada.ca; phone (416)
14 739-4840; FAX (416) 739-4281
15 *Author to whom correspondence should be addressed: Samar.Moussa@canada.ca; phone (416)
16 739-4942; FAX (416) 739-4281
18 **Supporting Information Available
19 Key words: Hydrogen cyanide, biodiesel, diesel, gasoline, emission factor, PFI, GDI
22 Abstract
23 Hydrogen cyanide (HCN) is considered a marker for biomass burning emissions and is a
24 component of vehicle exhaust. Despite its potential health impacts, vehicular HCN emissions
25 estimates and their contribution to regional budgets are highly uncertain. In the current study,
26 Proton Transfer Reaction-Time of Flight-Mass Spectrometry (PTR-ToF-MS) was used to
27 measure HCN emission factors from the exhaust of individual diesel, biodiesel and gasoline
28 vehicles. Laboratory emissions data as a function of fuel type and driving mode were combined
29 with ambient measurement data and model predictions. The results indicate that gasoline
30 vehicles have the highest emissions of HCN (relative to diesel fuel) and that biodiesel fuel has
31 the potential to significantly reduce HCN emissions even at realistic 5% blend levels. The data
32 further demonstrate that gasoline direct injection (GDI) engines emit more HCN than their port
33 fuel injection (PFI) counterparts, suggesting that the expected full transition of vehicle fleets to
34 GDI will increase HCN emissions. Ambient measurements of HCN in a traffic dominated area
35 of Toronto, Canada were strongly correlated to vehicle emission markers and consistent with
36 regional air quality model predictions of ambient air HCN, indicating that vehicle emissions of
37 HCN are the dominant source of exposure in urban areas. The results further indicate that
38 additional work is required to quantify HCN emissions from the modern vehicle fleet,
39 particularly in light of continuously changing engine, fuel and after-treatment technologies.
41 1. Introduction
42 Vehicular emissions are considered the dominant source of air pollution in urban areas
43 and have adverse effects on human health, air quality and the environment.1-6 Emissions from
44 vehicles depend upon vehicle type, age, maintenance, operation and fuels used, and contain a
45 complex mixture of particles and gaseous pollutants. Most vehicle emission testing has been
46 limited to common pollutants, such as carbon monoxide, nitrogen oxides and selected
47 hydrocarbons (including aromatic compounds) and more recently a range of particle-bound
48 compounds.7 However, given increased evidence of adverse health effects associated with
49 exposure to traffic-related air pollution ,6, 8 9 there is a need to identify and quantify emissions of
50 more pollutants which are potentially toxic and may be partly responsible for the observed health
51 impacts.5, 9 Such data will enable better estimates of traffic related air pollutant exposures and
52 assessments of health risks, and potentially help to identify more effective approaches to control
53 their emissions.
54 Hydrogen cyanide (HCN), along with acetonitrile (CH3CN), are the most abundant
55 cyanides in the atmosphere,10 yet the atmospheric budget of HCN remains uncertain. Due to its
56 stability and thus long lifetime in the atmosphere (~ 5 months), HCN has been considered a
57 biomass burning marker.11 Numerous satellite and aircraft measurements of HCN, 10, 12-26 and a
58 few reports from ground based locations,27, 28 suggest that on a global basis, biomass burning
59 (BB) emissions are a large source of HCN to the atmosphere (0.1-3.2 Tg-N-yr-1).12, 13, 17-19, 21, 24,
27, 29-33
61 Although limited measurements have reported HCN in the exhaust of vehicles 2 34-41, the
62 exact formation mechanism remains somewhat unclear. While some studies have indicated that
63 HCN is associated with the catalytic reduction of nitric oxides (NOx) over various catalytic
64 systems (selective catalytic reduction (SCR), silver, alumina, rhodium and platinum) typically
65 used to reduce NOx emissions. 35, 42, 43 Others have shown that HCN in exhaust is enhanced in the
66 absence of a catalyst.35 Despite the fact that HCN is a reduced species which can be created in
67 emission control systems (employing reduction catalysts) , the formation of HCN during biomass
68 burning suggests that HCN is also formed directly from fossil fuel combustion sources, which is
69 consistent with the work of Dagaut et al. 2008 who demonstrated that HCN is also an
70 intermediate in the formation of NOx in combustion via both prompt-NO and fuel-NO
71 mechanisms.44 Vehicle emission of HCN is also consistent with the reported variability in the
72 HCN correlation with CO and acetonitrile in biomass burning plumes in the vicinity of urban
73 areas, suggesting that urban sources of HCN are significant.12
74 Despite the studies reporting that HCN is emitted from vehicle emissions, no near-road
75 measurements of HCN have been performed and thus the contribution of vehicle emissions to
76 HCN population exposure in urban areas is unclear. Furthermore, with increased public concern
77 regarding air pollution, energy security, fuel efficiency and climate change, the use of alternative
78 fuels such as biodiesel and advanced engine technologies such as gasoline direct injection (GDI)
79 is increasing.45-50 While biodiesel fuels are generated from natural and renewable sources,49, 51-54
80 and have the potential to reduce emissions of atmospheric pollutants,46, 48, 49, 55-61 GDI engines
81 offer the promise of reduced CO2 emissions arising from increased fuel efficiency.50 However,
82 the potential effect of these evolving engine and fuel technologies on HCN emissions has not
83 been characterized. From a health perspective, quantifying urban HCN emissions is important,
84 given the known toxicity and the lifetime of this compound in the atmosphere. HCN is a toxic
85 compound that has adverse impacts on human health. While the acute effects for exposure to
86 HCN are known, the sub-chronic and chronic effects for exposures to ambient HCN levels are
87 poorly understood. Cyanide in the form of HCN and CN- at high levels (270 ppmv) is highly
88 toxic and can lead to death if inhaled or ingested.62-66 Exposure to HCN can contribute to acute
89 and chronic inflammatory lung diseases.66 In human blood, HCN is converted to thiocyanate
90 (SCN-) via the enzyme rhodanese in the presence of a sulfur donor group.64 A large number of
91 medical studies have documented the adverse effects of thiocyanate in humans, with most of
92 these studies having focused on cigarette smoking as the source of exposure to HCN.63-69
94 The emission of HCN from vehicles has been documented in a number of publications, the
95 majority of which are relatively old (1978-1983; 20 04-20 07).2, 34, 35, 37-40, 70 With the
96 improvement of vehicle technology with respect to engines, catalysts, ignition systems and
97 quality/types of fuel, the emission of HCN will likely have changed and hence HCN exhaust
98 emission factors from modern in-use vehicles would be useful for emission inventories, source
99 appointment and exposure studies. In the current study, emission factors (mg/kg of fuel
100 consumed) of HCN from various light duty vehicles (LDVs) are derived as a function of fuel
101 type (diesel, bio-diesel, gasoline) and driving mode, and with vehicles utilizing emerging engine
102 and exhaust after-treatment technologies. We demonstrate via a combination of laboratory
103 experiments, ambient measurements and air quality model predictions that vehicular emissions
104 are the dominant source of urban ambient HCN, representing a potentially important HCN
105 exposure pathway.
107 2. Experiments and Methods
108 2.1 PTR-TOF-MS
109 Measurements were performed using proton reaction time-of-flight mass spectrometry (PTR-
110 TOF-MS, Ionicon Analytik GmbH, Austria). The operating principles of the PTR-MS are
111 described in detail by Jordan et al. (2009) and Graus et al. (2010).71"72 In brief, the PTR-MS uses
112 a chemical ionization technique in which hydronium ions (H30+) are used as reagent ions to
113 protonate analyte molecules via Rl. The chemical ionization reaction by H30+ is considered a
114 soft process relative to electron impact, resulting in comparatively little fragmentation. A
115 number of papers have reported molecular fragmentation with the PTR-ToF -Ms depending on
116 the operating conditions and sampled matrix.5"73"77 Hydronium ions are generated by passing
117 water vapor through a hollow cathode ion source. The reagent ion and the sample gas enter a
118 drift tube, which was maintained at 2.15 mbar and at a temperature of 60 °C. The drift tube
119 operated at an E/N of 140 Td where species (X) with a proton affinity greater than that of water
120 (165 kcal/mol) are protonated according to reaction Rl.
121 X + H30+^ X.H++ H20 Rl
122 The product ions are then detected with a high resolution Time-of-Flight (HR-TOF) mass
123 spectrometer (mass resolution>3000; Tofwerk AG, Switzerland). Data analysis was performed
124 with the TOFWARE software package (Tofwerk AG). This software provides high resolution
125 peak fitting that is able to accurately integrate and separate the peaks from each other.
126 PTR-MS signals for HCN and other VOC species (benzene, toluene, xylenes and
127 acetonitrile) were calibrated with standard gas cylinders (HCN standard from Air Liquide at 5.5
128 ± 0.275 ppm in N2l VOC standards from IONIMED Analytik) through a gas calibration unit
129 (GCU, Ionicon, Analytik GmbH, Austria). Given that the proton affinity of HCN (170.4
130 kcal/mol) is only slightly greater than that of water, the PTR-TOF-MS response to HCN is
131 dependent on the relative humidity in the sample air stream28. To address this issue, relative
132 humidity dependent calibrations for HCN were performed and the resulting response factor-
133 relative humidity relationship was applied to subsequent data (Figure S1). The water
134 dependence of the response to HCN,28 H2S78 and HCHO79 in the PTR-MS has been reported
135 previously.
137 2.2 Engine Exhaust Experiments
138 2.2.1 Diesel and Biodiesel Engine Dynamometer Experiment
139 A turbo diesel injection (TDI), removed from a 2001 Volkswagen Jetta, was operated on an
140 engine dynamometer. The engine was equipped with a diesel oxidation catalyst (DOC) and was
141 operated on ultralow sulfur diesel (ULSD), three pure biodiesels (B100) reformulated from soy,
142 canola, and tallow/waste fry oils, respectively, and three blends (B5, B20, B50) of each biodiesel
143 with ULSD. B5, B20 and B50 represent the 5%, 20% and 50% by volume mixed blends with
144 ULSD, respectively. During the experiments, the engine was operated in four steady state
145 modes, including idle and three other conditions derived based upon average speed and engine
146 torque for the associated transient driving cycle: a) idle, b) US06 (Supplemental Federal Test
147 Procedure, aggressive driving mode), c) HWFET (Highway Fuel Economy Driving Schedule,
148 highway driving mode) and d) FTP75 (U.S. Federal Test Procedure 75, city driving mode). The
149 engine exhaust delivery and experimental setup have been described previously.80, 81 In brief, the
150 PTR-ToF-MS sampled from a constant volume sampler (CVS) dilution tunnel where HEPA
151 filtered and charcoal filtered room air was used to dilute the raw exhaust. The dilution factor
152 used in the CVS ranged from 15 to 80 depending upon the operation mode. In addition to the
153 PTR-Tof-MS measurements, concentrations of CO, CO2 (both with non-dispersive infrared
154 analyzers), NOx (chemiluminescence) and total hydrocarbons (THC, flame ionization detector)
155 were measured in the CVS (Table S1)
156 2.2.2 Gasoline Vehicle Chassis Dynamometer Studies
157 During this study, several gasoline light duty vehicles (LDVs) with different engine and
158 after-treatment technologies were configured on a chassis dynamometer for emission testing. The
159 vehicles were a 2011 Hyundai Sonata equipped with a wall-guided stoichiometric gasoline direct
160 injection (GDI) engine, a 2010 Volvo S40 and a 2008 Silverado light-duty truck (LDT) equipped
161 with a port fuel injection (PFI) engine. All vehicles were equipped with three way catalytic
162 converters (TWCs). The Sonata (GDI-LDV) and the Volvo (PFI-LDV) are certified as Tier 2
163 vehicles which operated on EPA Tier 2 certified gasoline. The transient driving cycles used in
164 this study were the US06, steady state 50 miles per hour (SS 50mph) and LA4 (urban driving).
165 A subset of the emission experiments were performed when the GDI vehicle was equipped with
166 a gasoline particulate filter (GPF). The GPF used in this study has been described in detail by
167 Chan et al.50, 82, 83 and was provided by the Manufactures of Emission Control Association
168 (MECA). Each test day began with two back to back LA4, followed by the 50 miles per hour
169 steady state, followed by two back to back US06 drive cycles. The raw exhaust from the vehicle
170 was directed and diluted using a full-flow CVS dilution tunnel, which is described in detail by
171 Chan et al. (2012, 2013).50, 83 Diluted exhaust was continuously sampled from the CVS over the
172 entire driving cycle by the PTR-ToF-MS. Concentrations of CO, CO2 (Horiba instruments AIA-
173 210 and AIA210LE, respectively), NOx (California Analytical Instruments, 400-HLCD) and
174 THC (California Analytical instruments, 300HFID) were also monitored (Table S2). Fuel-based
175 emission factors (EFs) were calculated using equation 1, where [A] and [CO2] are mass
176 concentrations in (ig/m3 for species of interest and CO2 respectively and Wc is the carbon mass
177 fraction of the fuel of interest. CO and VOCs are not included in the equation because they
178 account for <1% of the carbon in the fuel. HCN, VOCs and CO2 were background corrected
179 using zero air, prior to being used in the emission factors calculations
[A] MWCO
180 EF(mg / kgfuet) -O xWc (1)
[CO2] 12.01 c v ;
182 2.4 Ambient Measurements
183 Ambient measurements of HCN, acetonitrile, benzene, toluene and xylenes were
184 performed in Toronto, Ontario, Canada between June 28th and July 19th, 2013 at an urban
185 location approximately 100m from a major roadway. Another sampling period was conducted
186 between September 10th and September 17th, 2014 where the PTR-ToF-MS was deployed in the
187 Canadian Regional and Urban Investigation System for Environmental Research (CRUISER)
188 mobile laboratory. Latitude and longitude were measured using a GPS (GARMIN, GPSmap
189 176C). On most days driving started at approximately 10:00 am, continued through the
190 afternoon and ended between 5:00-7:00 PM for a total of 20 hours of driving over the study
191 period.
193 Ambient air in both cases was sampled through a perfluoroalkoxy (PFA) tube by a
194 diaphragm pump at a flow rate of 2-4 L min-1 of which 100 sccm was sampled by the PTR inlet
195 maintained at 60° C. Instrument background measurements were performed for 5 min every
196 hour using a platinum catalyst heated at 350 °C.
197 In addition to the PTR measurements in the 2013 campaign, parallel ambient
198 measurements were performed using a chemical ionization mass spectrometer (CIMS) with
199 iodide as the reagent ion 19, 81HCN was detected as CN- (m/z 26) and HCN.I- (m/z 154) using
200 iodide as the reagent ion as described elsewhere.19, 81.
202 2.5 Air Quality Emission Modelling
203 HCN predictions over Toronto for September 2014 (to coincide with mobile measurements),
204 were performed with Environment Canada's operational air quality model (GEM-MACH), 84, 85
205 to coincide with the mobile measurements. The air quality prediction system was run in an off-
206 line configuration with the meteorology driven by Environment Canada's operational weather
207 forecast model (GEM). For this study, GEM-MACH was run in a nested configuration down to
208 a 2.5-km grid spaced domain over Southern Ontario in which Toronto is located. HCN was
209 added to the modelling system as an emitted, transported and chemically reactive species, and
210 the model was run to test the sensitivity of HCN ambient levels to vehicle emission (on-road and
211 off-road gasoline and diesel). Other anthropogenic emissions (e.g. NOx, CO, VOCs) were based
212 on the 2008 US EPA and 2006 Canadian emission inventories (NPRI). The incorporation of
213 HCN and emission inventories in GEM-MACH is described in details in SI.
215 3. Results and Discussion
216 3.1 Gasoline Emission Studies
217 Figure 1 shows the dilution-corrected concentration profiles for benzene, toluene and HCN for
218 the duration of each transient driving mode for a GDI vehicle with no GPF. The Figure shows
219 that the concentrations of HCN, benzene, and toluene in the exhaust were 10 fold higher in the
220 full cold start cycle than when compared to the hot start cycle. HCN emissions from the exhaust
221 correlated with emissions of other traffic markers such as benzene and toluene and were on the
222 same order of magnitude as the aromatic compounds measured. Table 1 provides a summary of
223 the 22 minute average fuel-based emission factors for HCN, benzene, toluene and acetonitrile for
224 the different driving cycles. For the GDI vehicle, the fuel based emission factors for HCN were
225 2 times higher for cold start modes compared to the equivalent hot start mode. The highest
226 emission factor was obtained during the first US06 (aggressive) cycle followed by the LA4
227 (urban) cold drive cycle. The effect of engine and/ or catalyst temperature on HCN EFs was
228 even more pronounced in the PFI vehicle, where the EFs were 30 and 3 times higher for LA4
229 and US06 cold start compared to the equivalent hot starts, respectively.
230 A comparison of HCN emission factors (mg/km) from the exhaust of different vehicles
231 used in literature is shown in Table 2. There is clearly a wide range of HCN EFs reported in
232 literature which is due to the fact that exhaust emissions of HCN are known to vary greatly with
233 operation, engine, maintenance, age and control technology of the engine. Consequently, these
234 variables make direct comparison of HCN emissions from this study to other studies difficult.
235 This is further exacerbated by the limited number of vehicles used in this study (three gasoline
236 vehicles) compared to other studies, by the different driving cycles which were used here
237 compared to other studies, and by the fact that the three vehicles used in this study have newer
238 technology relative to others reported in the literature.
239 Despite the aforementioned caveats, the EFs obtained in this study fall within the same
240 range of some of the EFs reported in literature. For instance, Karlsson et al. (2004) was the only
241 study from Table 2 that reported emissions of HCN from an LA4 transient cycle. Qualitatively,
242 the average LA4 fuel-based EF for HCN (Table 1) for the GDI engine equipped with GPF is 6.65
243 mg/kg (~0.5 mg/km), which is on the same order of magnitude compared to the ~1.2 mg/km EF
244 reported by Karlsson et al. However, the average LA4 EF for the same vehicle not equipped
245 with a GPF is ~4.4 mg/km. Although no idle mode measurements for HCN from gasoline
246 vehicles were made in this study, the transient driving mode EFs suggest that HCN emissions
247 from non-idle driving modes are significant compared to idle mode values in other studies.34, 39
248 Despite the limited number of vehicles here, the effect of engine and/or catalyst
249 temperature on HCN emissions is consistent with the literature. Baum et al. (2007), reported
250 HCN emissions from 14 idling vehicles and they similarly showed that HCN emissions were
251 higher during cold engine starts compared to vehicles operating under temperature stabilized
252 conditions regardless of the make, year and model of the vehicle used.34 The increase in HCN
253 emissions during cold engine starts is also consistent with studies reporting the effect of catalyst
254 (and engine) temperature on the emissions of black carbon, THC, NOx and particulate matter
255 (PM) from GDI and PFI engines.50, 82, 83
256 A comparison of the calculated fuel-based emission factors for HCN from the GDI and the PFI
257 engines used in this study is shown in Figure 2A. The emissions of HCN from the GDI engine
258 ranged from 2-100 fold larger than those of the PFI engine, depending on the driving mode. An
259 increase in the emissions for the aromatic hydrocarbons was also observed in this study (Table
260 1). An increase in PM emissions from GDI vehicles relative to PFI vehicles has been reported in
50 82 83
261 other studies. 82, 83 To the best of our knowledge no studies have reported VOC emissions from
262 GDI vehicles relative to PFI vehicles. While a very large increase in HCN and VOCs emissions
263 from the GDI vehicle compared to the PFI was observed here, it is clear that further emission
264 studies with a larger sampling of vehicles are required to confirm this result.
265 The effect of a gasoline particulate filter (GPF) on the emissions of HCN from gasoline
266 vehicles was investigated on the GDI engine. Figure 2A shows fuel-based EFs of HCN from the
267 GDI engine with/without a GPF. The presence of the GPF decreased HCN emission by a factor
268 of 10, which may be due to the adsorption of HCN/cyanide on the metal filter surface. The GPF
269 used in this study was made from cordierite (magnesium iron aluminum cyclosilicate) material.
270 Yi et al. have shown that HCN, acetonitrile and benzene decrease significantly on such metal
271 surfaces.86 Additionally, a number of studies have shown that CN- can bind efficiently to metal
272 sites as an intermediate in the reduction of NOx on different metal catalysts.87-91
274 3.2 Diesel and Biodiesel Emission Studies
275 A summary of the calculated fuel based EFs for HCN from ULSD, and pure biodiesel from
276 (B100) tallow/waste fry oil, soy and canola for various steady states is shown in Figure 2, and
277 represents the first known report of such values. The average HCN fuel based EFs for the four
278 steady state driving modes (HWFET, FTP75, US06 and idle) from ULSD were 3.0 ± 0.4, 3.9 ±
279 0.5, 3.4 ± 0.4 and 10.3 ± 2.8 mg kg Fuel-1, respectively. The fuel based EFs for HCN from
280 ULSD were ~10 times lower than those obtained from the PFI and GDI gasoline vehicles during
281 US06 driving mode. However, it is important to note that integrating a transient drive cycle is
282 typically not equivalent to the steady state tests, because emissions are expected to increase with
283 the aggressiveness and transient nature of the driving condition. A further and significant
284 reduction in HCN emissions from the engine was achieved (relative to ULSD) when operating
285 on the biodiesel fuels (Figure 2B, 2C). The reduction ranged from 20-80% depending on the
286 operating mode, where the largest reduction in HCN emissions (~80%) for all the biofuels was
287 achieved in the idle mode. Intriguingly, the same reduction in HCN emission from pure
288 biodiesels (B100) was achieved using B05, B20 and B50 blends (Figure 2 and Figure S2). This
289 may be related to the mechanism by which HCN (an intermediate product) is converted to NOx
290 during combustion. Further studies are needed to determine the exact formation mechanism for
291 HCN which is beyond the scope of this study. The reduction in HCN EFs with biodiesel fuel use
292 is also concurrent with a reduction in aromatic hydrocarbon emissions as shown in Table 3, and
293 consistent with hydrocarbon emission reductions reported with biofuel use in a number of
294 studies.57, 58, 61, 92-94
296 3.3 Ambient Measurements and Modeling
297 To further demonstrate the contribution of HCN vehicle emissions to air quality, near-
298 road ambient measurements were performed in Toronto, Canada. The diurnal profiles of HCN,
299 acetonitrile (ACN), toluene and benzene concentrations in ambient air during 2013 are shown in
300 Figure 3. An increase in the HCN concentration was observed during the morning rush hour
301 period (~ 6:00 - 8:00 am), which coincided with the increase in other traffic markers such as
302 benzene and toluene, but not with acetonitrile. A positive correlation observed between HCN
303 and benzene during ambient measurements (Figure S3), strongly suggests that ambient HCN is,
304 at least in large part, associated with vehicle emissions. 81, 95, The diurnal profile of ambient
305 HCN is consistent with the air quality model predictions in the region (Figure 4) which shows
306 that HCN emissions are higher during morning periods (7:00-10:00 AM) relative to midday and
307 evening. Conversely, there was a poor correlation between CH3CN and HCN during the
308 sampling period despite both species having been used as biomass burning markers and
309 measured in vehicle exhaust (Tables 1 and 3), suggesting different sources for both species.12
310 Such an observation may be expected since the fuel-based emission factor for CH3CN from
311 gasoline and diesel emissions were minimal compared to HCN (Tables 1 and 3). This is also
312 consistent with the report that vehicle emissions are a minor source of atmospheric CH3CN,96
313 and with measurements over Central Mexico, where an increase in ACN levels was observed
314 without an accompanying enhancement in HCN. 12 The HCN PTR measurements during 2013
315 are supported by parallel measurements conducted by the CIMS for the same period. A similar
316 HCN profile was obtained using the CIMS (Figure S4).
318 During the 2013 sampling period, HCN levels ranged from ~ 0.2 ppb to 20.0 ppb with a mean
319 value of 3.45 ± 3.43 ppb (1o), and a median of 2.17 ppb, respectively. To investigate the spatial
320 extent and distribution of HCN in an urban region affected by significant vehicle emissions, a
321 second sampling period was conducted in September 2014 coinciding with the output of an air
322 quality model. HCN levels during this study, which was able to examine a greater number of
323 locations due to use of the mobile lab, ranged from ~ 0.1 ppb to 3.8 ppb with a mean value of
324 1.57 ± 0.33 ppbb (1 o) (Figure S5). A direct comparison between the two sampling periods is not
325 straight forward since continuous sampling was conducted for 2 weeks in 2013 at one location
326 compared to intermittent measurements (3 - 4 hrs per day) in different locations around the city
327 during the mobile measurements in 2014. The HCN levels (Figure S5) from 2014 are in
328 agreement with the GEM-MACH midday and evening predictions (Figure 4) during the same
329 period, suggesting that HCN emissions from vehicles are likely the major source of HCN in this
330 region, since the only source of HCN in the model was from vehicle emissions. HCN levels
331 during the September 2013 measurement period were generally higher than during the mobile
332 study, with a regional background of ~ 2 ppb. This may reflect an influence from regional
333 biomass burning episodes during the 2013 measurement period. There were episodes of biomass
334 burning in parts of the province of Quebec during the sampling period; however, backward wind
335 trajectories calculated using NOAA-Hybrid Single-Particle Lagrangian Integrated Trajectory
336 Model (HYSPLIT) indicated that the biomass burning plume did not impact the near-road
337 location.
338 Measurements of HCN during both sampling periods were significantly higher than the
339 mean HCN mixing ratios of ~ 0.2 ppb and 0.36 ± 0.16 ppb reported by Knighton et al. (2009)
340 and Ambrose et al. (2012), respectively. , Such differences may be attributed to differences in
341 the sampling locations. Specifically, during 2013, the measurements were made about 100 m
342 from a heavily commuted street in Toronto whereas the study of Ambrose et al., 2012 was
343 conducted near a hardwood/pine forest. Additionally, the strong correlation between HCN and
344 CH3CN in Ambrose et al. (2012) was more consistent with a large scale, aged biomass burning
345 influence rather than vehicle emissions.27 Knighton et al. (2009), sampled at a suburban site in
346 Boston; however, their HCN measurements were likely influenced by interferences from C2H4+
347 in the quadrupole PTR-MS signal which could not be resolved from HCN at m/z 28,28 whereas
348 both ions were resolvable in the current study with the HR-TOF mass spectrometer. In the PTR-
349 TOF the signal from HCN.H+ and C2H4+ are detected at m/z 28.01818 and m/z 28.0308,
350 respectively. However in the PTR-quadrupole MS, both signals will be observed at nominal m/z
351 28. Attempts at subtracting the contribution of C2H4+ from the signal at m/z 28 in that study will
352 increase the uncertainty in the derived HCN. The aforementioned studies represent the most
10 22 97
353 relevant comparisons , , for HCN in the urban ambient air, since most other studies have
354 been performed in the middle to upper troposphere and mostly for large scale biomass burning
355 estimates.
359 3.4 HCN Emissions Estimates for Canada
360 The strong correlation between HCN and aromatic compounds (Figure S3) during ambient
361 measurements and the fuel-based EFs of HCN from the gasoline and diesel engine laboratory
362 studies clearly suggest that vehicle emissions can be a source of HCN in urban areas. Despite
363 the caveats discussed in the previous section, and to put our results in perspective, we assumed
364 that the on-road vehicle fleet has HCN EFs similar to the values reported in this study. In that
365 case, the annual emissions of HCN for Ontario and Canada can be estimated using the fuel based
366 emission factors obtained from the current laboratory study and from the annual sales of gasoline
367 (~99%) and diesel (1%) for LDV reported by Statistics Canada.98 A summary of total annual
368 HCN emissions estimates for Ontario and Canada is provided in Table 4. Approximately 40 x
369 106 thousand liters of gasoline were sold in Canada nationwide in 2012, of which 38.5% were
370 sold in the province of Ontario (Table 4). Using an average fuel-based emission factor for HCN
371 (21.0 mg/kg of gasoline, Table 1) and 0.77 kg/L as the density of gasoline, we estimate that 654
372 tonnes of HCN were emitted in 2012 from LDV in Canada, including ~252 tonnes in Ontario.
373 The same approach can be used to estimate HCN emissions from LDVS operating on ULSD
374 which results in 14 and 49 tonnes of HCN emitted in 2012 from vehicles for Ontario and Canada,
375 respectively (Table 4). The average emission factor from the current diesel study is 3.2 mg/ kg
376 of ULSD (Table 3, excluding idle mode) and the density of diesel used was ~0.87 kg/L.
377 On a national scale, we estimate that 703 tonnes of HCN are emitted annually from the
378 combined gasoline and diesel LDV fleets. However, the current estimate likely represents an
379 underestimate of the total HCN emissions as it does not account for the contribution of heavy
380 duty diesel vehicles (HDDV), is based upon a sampling of three vehicles only, and does not
381 account for the variability in HCN EFs. The 703 tonnes is further considered an underestimate
382 since there are several studies which suggest that poor maintenance of vehicles and
383 malfunctioning catalysts enhances HCN emissions (as discussed in more details in section 3.1) 38,
384 40 whereas the vehicles in the current study were relatively new and well maintained.
385 To place vehicle emissions of HCN in the context of the total Canadian HCN budget
386 (which includes biomass burning emissions of HCN) the emission ratio (ER) of HCN/CO for
387 forest fires is used to estimate the contribution of HCN from biomass burning nationwide. These
388 estimates are also summarized in Table 4 and described in detail in SI. The estimates in Table 2
389 indicate that despite the large uncertainties and the number of assumptions involved in biomass
390 burning estimates for HCN, national emissions from biomass burning are 3-321 times higher
391 than HCN emissions from vehicles. It is estimated that 70-85% of the global HCN budget is
392 due to biomass burning emissions,21, 31 therefore HCN vehicle emissions (703 tonnes) may be
393 considered non-negligible. However, this likely represents an important source of population
394 exposure to HCN in urban areas particularly since forest fire plumes are episodic in nature while
395 vehicle emissions of HCN are persistent and concentrated near the ground in close proximity to
396 population. Clearly, more emission studies of HCN from a bigger sampling of vehicles are
397 required.
398 4. Conclusions and Implications
399 The results of the present combined laboratory exhaust emissions, modelling and ambient
400 sampling study reports HCN emissions from different types of vehicles and fuels under a range
401 of standard driving modes. The results indicate that gasoline vehicle emissions under transient
402 driving conditions are likely the dominant source of ambient HCN in urban areas. This is
403 important since a large fraction of urban population is continuously exposed to vehicle
404 emissions, either by being inside a vehicle or by living and working in the vicinity of traffic,
405 including major roadways. 5 6 9
406 The shift from PFI to GDI vehicles is occurring rapidly due to the promise of increased
407 fuel efficiency and reduced CO2 emissions. It is expected that by 2025, GDI vehicles will
408 dominate the market (~93%), replacing less efficient PFI vehicles.99 Our results demonstrate
409 that, moving fully towards GDI vehicles, will not only increase HCN emissions but it will also
410 increase the emissions of other toxic VOCs such as benzene and xylene. This is intriguing and
411 calls for more emission studies that include a bigger and more representative sampling of on-
412 road PFI and GDI vehicles to investigate the consequences of shifting from PFI to GDI vehicles
413 on air quality. While GPF may provide future GDI vehicles a means to reduce ultrafine particle
49, 76, 77
414 emissions, , , our data suggest that there may also be an unexpected co-benefit of lower HCN
415 emissions with the use of a GPF. However, additional work will be required to confirm not only
416 this reduction but also that it is maintained as the GPF ages.
417 The current data also indicates that HCN emissions are significantly reduced when using
418 biodiesel blends relative to ULSD. Note that the gasoline HCN emissions were obtained from
419 transient driving conditions while diesel HCN emissions were measured during steady state.
420 Consequently a direct comparison between gasoline and diesel EFs is not quantitatively possible.
421 However, considering that 99% of the diesel fuel sold in Canada is used by heavy duty vehicles
422 (HDVs; not LDDVs), in a relative sense, the total HCN emissions contributed by all light-duty
423 gasoline vehicles are likely much higher than all HCN emissions from light-duty diesel vehicles.
424 Additional data is required to assess effect of biodiesel on the emissions of HCN from heavy-
425 duty vehicles.
426 From the human health perspective, there is increased evidence that traffic related air
427 pollution (TRAP) is linked to cardiovascular and respiratory diseases, cancer, and increased
428 mortality and morbidity rates.6, 81 While the acute effects for exposure to high levels of HCN are
429 known, the acute and chronic effects of low exposures (i.e., at the ambient levels reported here)
430 are poorly understood.
431 There is no federal air quality standard for HCN in Canada; however, based upon the
432 inhalation reference concentration (RfC) reported in an occupational study by El Ghawabi et al.
433 (1975),100 the province of Ontario has established an air quality standard (AAQC) for HCN. The
434 24 hr average and half hour point impingement AAQC for HCN in Ontario are 8 (ig/m3 (~ 7ppb)
435 and 24 (ig/m3 (~ 20 ppb), respectively.101 During the ambient measurements of this study in
436 Toronto, Canada's largest city and the third largest in North America, the average HCN
437 concentration in the 2013 ambient data was 3.34 ppb. During the 19 day sampling period in
438 2013, the 24 hr average exceeded the Ontario AAQC and the EPA RfC value (0.003 mg/m3 or
439 ~3 ppb)62 on 3 and 5 days, respectively (Figure S6). Model predictions (Figure 4C) also
440 demonstrate that HCN levels (derived from vehicle emissions) can reach ~ 5 ppb during the
441 morning rush hour in downtown Toronto and on major highways.
442 This study suggests that there is a need for more emission studies and new
443 health/exposure data characterizing the effect of prolonged exposure to ambient levels of HCN,
444 particularly in light of the development of new vehicle technologies and after-treatment devices ,
445 which have the potential to increase HCN emissions and other traffic related pollutants.
448 Acknowledgements
449 The authors would like to acknowledge the financial support for this work through the Particles
450 and Related Emission Project (PERD) C11.008, which is a program administered by Natural
451 Resources Canada (NRC). We also thank Dr. Hans Osthoff, Elisabeth Galarneau, Yongchun Liu,
452 Ali Abdul Sater, Ralf Stablear, Rob McLaren and Alex Lupu, Andrea Darlington and Richard
453 Mettimier for useful discussions. We thank David Niemi for providing NPRI emission
454 estimates. The authors thank Joe Kubsh and Rasto Brezny from MECA for providing the GPF
455 used in the gasoline study.
457 References
459 1. Franco, V.; Kousoulidou, M.; Muntean, M.; Ntziachristos, L.; Hausberger, S.; Dilara, P.,
460 Road vehicle emission factors development: A review. Atmos. Environ. 2013, 70, (0), 84-97.
461 2. Karlsson, H. L., Ammonia, nitrous oxide and hydrogen cyanide emissions from five
462 passenger vehicles. Sci. Total Environ. 2004, 334-335, (0), 125-132.
463 3. May, A. A.; Nguyen, N. T.; Presto, A. A.; Gordon, T. D.; Lipsky, E. M.; Karve, M.;
464 Gutierrez, A.; Robertson, W. H.; Zhang, M.; Brandow, C.; Chang, O.; Chen, S.; Cicero-
465 Fernandez, P.; Dinkins, L.; Fuentes, M.; Huang, S.-M.; Ling, R.; Long, J.; Maddox, C.; Massetti,
466 J.; McCauley, E.; Miguel, A.; Na, K.; Ong, R.; Pang, Y.; Rieger, P.; Sax, T.; Truong, T.; Vo, T.;
467 Chattopadhyay, S.; Maldonado, H.; Maricq, M. M.; Robinson, A. L., Gas- and particle-phase
468 primary emissions from in-use, on-road gasoline and diesel vehicles. Atmos. Environ. 2014, 88,
469 (0), 247-260.
470 4. Vardoulakis, S.; Fisher, B. E. A.; Pericleous, K.; Gonzalez-Flesca, N., Modelling air
471 quality in street canyons: a review. Atmos. Environ. 2003, 37, (2), 155-182.
472 5. HEI Panel on the Health Effects of Traffic Related Air Polltion. 2010. Traffic Related Air
473 Pollution : A Critical Review of the Literature on Emissions, Exposure and Health Effects. HEI
474 Special Reportl 7. Health Effects Institute, Boston, MA.
475 6. Jerrett, M.; Finkelstein, M. M.; Brook, J. R.; Arain, M. A.; Kanaroglou, P.; Stieb, D. M.;
476 Gilbert, N. L.; Verma, D.; Finkelstein, N.; Chapman, K. R., A cohort study of traffic-related air
477 pollution and mortality in Toronto, Ontario, Canada. Environ. Health Persp. 2009, 117, (5), 772478 7.
479 7. Zielinska, B.; Sagebiel, J.; McDonald, J. D.; Whitney, K.; Lawson, D. R., Emission Rates
480 and Comparative Chemical Composition from Selected In-Use Diesel and Gasoline-Fueled
481 Vehicles. J. Air Waste Manage. Assoc. 2004, 54, (9), 1138-1150.
482 8. Beckerman, B.; Jerrett, M.; Brook, J. R.; Verma, D. K.; Arain, M. A.; Finkelstein, M. M.,
483 Correlation of nitrogen dioxide with other traffic pollutants near a major expressway. Atmos.
484 Environ. 2008, 42, (2), 275-290.
485 9. McKeown, D. Air Pollution Burden of Illness from Traffic in Toronto Problems and
486 Solutions; Toronto Publinc Health: Toronto, 2007.
487 10. Singh, H. B.; Salas, L.; Herlth, D.; Kolyer, R.; Czech, E.; Viezee, W.; Li, Q.; Jacob, D. J.;
488 Blake, D.; Sachse, G.; Harward, C. N.; Fuelberg, H.; Kiley, C. M.; Zhao, Y.; Kondo, Y., In situ
489 measurements of HCN and CH3CN over the Pacific Ocean: Sources, sinks, and budgets. J.
490 Geophys. Res.-Atmos. 2003, 108, (D20), 8795.
491 11. Bunkan, A. J. C.; Liang, C.-H.; Pilling, M. J.; Nielsen, C. J., Theoretical and
492 experimental study of the OH radical reaction with HCN. Mol. Phys. 2013, 111, (9-11), 1589493 1598.
494 12. Crounse, J. D.; DeCarlo, P. F.; Blake, D. R.; Emmons, L. K.; Campos, T. L.; Apel, E. C.;
495 Clarke, A. D.; Weinheimer, A. J.; McCabe, D. C.; Yokelson, R. J.; Jimenez, J. L.; Wennberg, P.
496 O., Biomass burning and urban air pollution over the Central Mexican Plateau. Atmos. Chem.
497 Phys. 2009, 9, (14), 4929-4944.
498 13. Glatthor, N.; Höpfner, M.; Semeniuk, K.; Lupu, A.; Palmer, P. I.; McConnell, J. C.;
499 Kaminski, J. W.; von Clarmann, T.; Stiller, G. P.; Funke, B.; Kellmann, S.; Linden, A.; Wiegele,
500 A., The Australian bushfires of February 2009: MIPAS observations and GEM-AQ model
501 results. Atmos. Chem. Phys. 2013, 13, (3), 1637-1658.
502 14. Goode, J. G.; Yokelson, R. J.; Ward, D. E.; Susott, R. A.; Babbitt, R. E.; Davies, M. A.;
503 Hao, W. M., Measurements of excess O3, CO2, CO, CH4, C2H4, C2H2, HCN, NO, NH3,
504 HCOOH, CH3COOH, HCHO, and CH3OH in 1997 Alaskan biomass burning plumes by
505 airborne Fourier transform infrared spectroscopy (AFTIR). J. Geophys. Res.-Atmos. 2000, 105,
506 (D17), 22147-22166.
507 15. Shim, C.; Wang, Y.; Singh, H. B.; Blake, D. R.; Guenther, A. B., Source characteristics
508 of oxygenated volatile organic compounds and hydrogen cyanide. J. Geophys. Res.-Atmos. 2007,
509 112, (D10), D10305.
510 16. Simpson, I. J.; Akagi, S. K.; Barletta, B.; Blake, N. J.; Choi, Y.; Diskin, G. S.; Fried, A.;
511 Fuelberg, H. E.; Meinardi, S.; Rowland, F. S.; Vay, S. A.; Weinheimer, A. J.; Wennberg, P. O.;
512 Wiebring, P.; Wisthaler, A.; Yang, M.; Yokelson, R. J.; Blake, D. R., Boreal forest fire emissions
513 in fresh Canadian smoke plumes: C1-C10 volatile organic compounds (VOCs), CO2, CO, NO2,
514 NO, HCN and CH3CN. Atmos. Chem. Phys. 2011, 11, (13), 6445-6463.
515 17. Viatte, C.; Strong, K.; Paton-Walsh, C.; Mendonca, J.; O'Neill, N. T.; Drummond, J. R.,
516 Measurements of CO, HCN, and C2H6 Total Columns in Smoke Plumes Transported from the
517 2010 Russian Boreal Forest Fires to the Canadian High Arctic. Atmos. Ocean 2013, 51, (5), 522518 531.
519 18. Duflot, V.; Hurtmans, D.; Clarisse, L.; R'honi, Y.; Vigouroux, C.; De Maziere, M.;
520 Mahieu, E.; Servais, C.; Clerbaux, C.; Coheur, P.-F., Measurements of hydrogen cyanide (HCN)
521 and acetylene (C2H2) from the Infrared Atmospheric Sounding Interferometer (IASI). Atmos.
522 Meas. Tech. 2013, 6.
523 19. Le Breton, M.; Bacak, A.; Muller, J. B. A.; O'Shea, S. J.; Xiao, P.; Ashfold, M. N. R.;
524 Cooke, M. C.; Batt, R.; Shallcross, D. E.; Oram, D. E.; Forster, G.; Bauguitte, S. J. B.; Palmer, P.
525 I.; Parrington, M.; Lewis, A. C.; Lee, J. D.; Percival, C. J., Airborne hydrogen cyanide
526 measurements using a chemical ionisation mass spectrometer for the plume identification of
527 biomass burning forest fires. Atmos. Chem. Phys. 2013, 13, (18), 9217-9232.
528 20. Liang, Q.; Jaegle, L.; Hudman, R. C.; Turquety, S.; Jacob, D. J.; Avery, M. A.; Browell,
529 E. V.; Sachse, G. W.; Blake, D. R.; Brune, W.; Ren, X.; Cohen, R. C.; Dibb, J. E.; Fried, A.;
530 Fuelberg, H.; Porter, M.; Heikes, B. G.; Huey, G.; Singh, H. B.; Wennberg, P. O., Summertime
531 influence of Asian pollution in the free troposphere over North America. J. Geophys. Res.-
532 Atmos. 2007, 112, (D12), D12S11.
533 21. O'Shea, S. J.; Allen, G.; Gallagher, M. W.; Bauguitte, S. J. B.; Illingworth, S. M.; Le
534 Breton, M.; Muller, J. B. A.; Percival, C. J.; Archibald, A. T.; Oram, D. E.; Parrington, M.;
535 Palmer, P. I.; Lewis, A. C., Airborne observations of trace gases over boreal Canada during
536 BORTAS: campaign climatology, air mass analysis and enhancement ratios. Atmos. Chem. Phys.
537 2013, 13, (24), 12451-12467.
538 22. Rinsland, C. P.; Dufour, G.; Boone, C. D.; Bernath, P. F.; Chiou, L.; Coheur, P.-F.;
539 Turquety, S.; Clerbaux, C., Satellite boreal measurements over Alaska and Canada during June-
540 July 2004: Simultaneous measurements of upper tropospheric CO, C2H6, HCN, CH3Cl, CH4,
541 C2H2, CH3OH, HCOOH, OCS, and SF6 mixing ratios. Global Biogeochem. Cycles 2007, 21,
542 (3), GB3008.
543 23. Rinsland, C. P.; Goldman, A.; Murcray, F. J.; Stephen, T. M.; Pougatchev, N. S.;
544 Fishman, J.; David, S. J.; Blatherwick, R. D.; Novelli, P. C.; Jones, N. B.; Connor, B. J., Infrared
545 solar spectroscopic measurements of free tropospheric CO, C2H6, and HCN above Mauna Loa,
546 Hawaii: Seasonal variations and evidence for enhanced emissions from the Southeast Asian
547 tropical fires of 1997-1998. J. Geophys. Res.-Atmos. 1999, 104, (D15), 18667-18680.
548 24. Rinsland, C. P.; Mahieu, E.; Zander, R.; Demoulin, P.; Forrer, J.; Buchmann, B., Free
549 tropospheric CO, C2H6, and HCN above central Europe: Recent measurements from the
550 Jungfraujoch station including the detection of elevated columns during 1998. J. Geophys. Res.-
551 Atmos. 2000, 105, (D19), 24235-24249.
552 25. Vigouroux, C.; Stavrakou, T.; Whaley, C.; Dils, B.; Duflot, V.; Hermans, C.; Kumps, N.;
553 Metzger, J. M.; Scolas, F.; Vanhaelewyn, G.; Müller, J. F.; Jones, D. B. A.; Li, Q.; De Maziere,
554 M., FTIR time-series of biomass burning products (HCN, C2H6, C2H2, CH3OH, and HCOOH)
555 at Reunion Island (21° S, 55° E) and comparisons with model data. Atmos. Chem.
556 Phys. 2012, 12, (21), 10367-10385.
557 26. Wiegele, A.; Glatthor, N.; Höpfner, M.; Grabowski, U.; Kellmann, S.; Linden, A.; Stiller,
558 G.; von Clarmann, T., Global distributions of C2H6, C2H2, HCN, and PAN retrieved from
559 MIPAS reduced spectral resolution measurements. Atmos. Meas. Tech. 2012, 5, (4), 723-734.
560 27. Ambrose, J. L.; Zhou, Y.; Haase, K.; Mayne, H. R.; Talbot, R.; Sive, B. C., A gas
561 chromatographic instrument for measurement of hydrogen cyanide in the lower atmosphere.
562 Atmos. Meas. Tech. 2012, 5, (6), 1229-1240.
563 28. Knighton, W. B.; Fortner, E. C.; Midey, A. J.; Viggiano, A. A.; Herndon, S. C.; Wood, E.
564 C.; Kolb, C. E., HCN detection with a proton transfer reaction mass spectrometer. Int. J. Mass
565 spectrom. 2009, 283, (1-3), 112-121.
566 29. Glatthor, N.; von Clarmann, T.; Stiller, G. P.; Funke, B.; Koukouli, M. E.; Fischer, H.;
567 Grabowski, U.; Höpfner, M.; Kellmann, S.; Linden, A., Large-scale upper tropospheric pollution
568 observed by MIPAS HCN and C2H6 global distributions. Atmos. Chem. Phys. 2009, 9, (24),
569 9619-9634.
570 30. Li, Q.; Jacob, D. J.; Bey, I.; Yantosca, R. M.; Zhao, Y.; Kondo, Y.; Notholt, J.,
571 Atmospheric hydrogen cyanide (HCN): Biomass burning source, ocean sink? Geophys. Res. Lett.
572 2000, 27, (3), 357-360.
573 31. Li, Q.; Jacob, D. J.; Yantosca, R. M.; Heald, C. L.; Singh, H. B.; Koike, M.; Zhao, Y.;
574 Sachse, G. W.; Streets, D. G., A global three-dimensional model analysis of the atmospheric
575 budgets of HCN and CH3CN: Constraints from aircraft and ground measurements. J. Geophys.
576 Res.-Atmos. 2003, 108, (D21), 8827.
577 32. Li, Q.; Palmer, P. I.; Pumphrey, H. C.; Bernath, P.; Mahieu, E., What drives the observed
578 variability of HCN in the troposphere and lower stratosphere? Atmos. Chem. Phys. 2009, 9, (21),
579 8531-8543.
580 33. Lupu, A.; Kaminski, J. W.; Neary, L.; McConnell, J. C.; Toyota, K.; Rinsland, C. P.;
581 Bernath, P. F.; Walker, K. A.; Boone, C. D.; Nagahama, Y.; Suzuki, K., Hydrogen cyanide in the
582 upper troposphere: GEM-AQ simulation and comparison with ACE-FTS observations. Atmos.
583 Chem. Phys. 2009, 9, (13), 4301-4313.
584 34. Baum, M. M.; Moss, J. A.; Pastel, S. H.; Poskrebyshev, G. A., Hydrogen Cyanide
585 Exhaust Emissions from In-Use Motor Vehicles. Environ. Sci. Technol. 2006, 41, (3), 857-862.
586 35. Keirns, M. H.; Holt, E. L., Hydrogen cyanide emissions from three-way catalyst
587 prototypes under malfunctioning conditions. SAE Tech. Pap. Ser. 1978, 780201.
588 36. Bradow, R. L.; Stump, F. D., Unregulated emissions from three-way catalyst cars. SAE
589 Tech. Pap. Ser. 1977, 770369.
590 37. Cadle, S. H.; Nebel, G. J.; Williams, R. L., Measurements of unregulated emissions from
591 General Motors' light-duty vehicles. SAE Tech. Pap. Ser. 1979, 790694.
592 38. Urban, C. M.; Garbe, R. J., Regulated and unregulated exhaust emissions from
593 malfunctioning automobiles. SAE Tech. Pap. Ser. 1979, 790696.
594 39. Harvey, C. A.; Garbe, R. J.; Baines, T. M.; Somers, J. H.; Hellman, K. H.; Carey, P. M.,
595 A study of the impact of some unregulated motor vehicle emissions. SAE Tech. Pap. Ser. 1983,
596 830987.
597 40. Urban, C. M.; Garbe, R. J., Exhaust emissions form malfunctioning three-way catalyst-
598 equipped automobiles. SAE Tech. Pap. Ser. 1980, 800511.
599 41. Becker, K. H.; Lorzer; Kurtenbach, R.; Wiesen, P.; Jensen, T. E.; Wallington, T. J.,
600 Nitrous Oxide (N2O) Emissions from Vehicles. Environ. Sci. Technol. 1999, 33, (22), 4134601 4139.
602 42. Tamm, S.; Ingelsten, H. H.; Palmqvist, A. E. C., On the different roles of isocyanate and
603 cyanide species in propene-SCR over silver/alumina. J. Catal. 2008, 255, (2), 304-312.
604 43. Voorhoeve, R. J. H.; Patel, C. K. N.; Trimble, L. E.; Kerl, R. J., Hydrogen cyanide
605 production during reduction of nitric oxide over platinum catalysts. Science 1975, 190, 149-151.
606 44. Dagaut, P.; Glarborg, P.; Alzueta, M. U., The oxidation of hydrogen cyanide and related
607 chemistry. Prog. Energy Combust. Sci. 2008, 34, (1), 1-46.
608 45. Basha, S. A.; Gopal, K. R.; Jebaraj, S., A review on biodiesel production, combustion,
609 emissions and performance. Renew. Sust. Energ. Rev. 2009, 13, (6-7), 1628-1634.
610 46. Di, Y.; Cheung, C. S.; Huang, Z., Comparison of the effect of biodiesel-diesel and
611 ethanol-diesel on the gaseous emission of a direct-injection diesel engine. Atmos. Environ. 2009,
612 43, (17), 2721-2730.
613 47. Fernando, S.; Hall, C.; Jha, S., NOx Reduction from Biodiesel Fuels. Energ. Fuel 2005,
614 20, (1), 376-382.
615 48. Shahid, E. M.; Jamal, Y., A review of biodiesel as vehicular fuel. Renew. Sust. Energ.
616 Rev. 2008, 12, 2484-2494.
617 49. Wang, W. G.; Lyons, D. W.; Clark, N. N.; Gautam, M.; Norton, P. M., Emissions from
618 Nine Heavy Trucks Fueled by Diesel and Biodiesel Blend without Engine Modification.
619 Environ. Sci. Technol. 2000, 34, (6), 933-939.
620 50. Chan, T. W.; Meloche, E.; Kubsh, J.; Brezny, R.; Rosenblatt, D.; Rideout, G., Impact of
621 ambient temperature on gaseous and particle emissions from a direct injection gasoline vehicle
622 and its implications on particle filtration. SAE Int. J. Fuels Lubr. 2013, 6, (2), 350-371.
623 51. Pinto, A. C.; Guarieiro, L. L. N.; Rezende, M. J. C.; Ribeiro, N. M.; Torres, E. A.; Lopes,
624 W. A.; Pereira, P. A. d. P.; Andrade, J. B. d., Biodiesel: an overview. J. Brazil. Chem. Soc. 2005,
625 16, 1313-1330.
626 52. Lapuerta, M.; Armas, O.; Ballesteros, R.; Fernandez, J., Diesel emissions from biodiesel
627 derived from Spanish potential vegetable oils. Fuel 2005, 84, 773-780.
628 53. Sun, J.; Caton, J. A.; Jacobs, T. J., Oxides of nitrogen emissions from biodiesel-fuelled
629 diesel engines. Prog. Energy Combust. Sci. 2010, 36, (6), 677-695.
630 54. Anderson, L. G., Effects of biodiesel fuels use on vehicle emissions. JSEE 2012, 3, 35631 47.
632 55. Bakeas, E.; Karavalakis, G.; Stournas, S., Biodiesel emissions profile in modern diesel
633 vehicles. Part 1: Effect of biodiesel origin on the criteria emissions. Sci. Total Environ. 2011,
634 409, (9), 1670-1676.
635 56. Basha, S. A.; Raja Gopal, K., A review of the effects of catalyst and additive on biodiesel
636 production, performance, combustion and emission characteristics. Renew. Sust. Energ. Rev.
637 2012, 16, (1), 711-717.
638 57. Di, Y.; Cheung, C. S.; Huang, Z., Experimental investigation on regulated and
639 unregulated emissions of a diesel engine fueled with ultra-low sulfur diesel fuel blended with
640 biodiesel from waste cooking oil. Sci. Total Environ. 2009, 407, (2), 835-846.
641 58. Peng, C.-Y.; Yang, H.-H.; Lan, C.-H.; Chien, S.-M., Effects of the biodiesel blend fuel on
642 aldehyde emissions from diesel engine exhaust. Atmos. Environ. 2008, 42, (5), 906-915.
643 59. Zou, L.; Atkinson, S., Characterising vehicle emissions from the burning of biodiesel
644 made from vegetable oil. Environ. Technol. 2003, 24, (10), 1253-60.
645 60. Karavalakis, G.; Stournas, S.; Bakeas, E., Effects of diesel/biodiesel blends on regulated
646 and unregulated pollutants from a passenger vehicle operated over the European and the Athens
647 driving cycles. Atmos. Environ. 2009, 43, (10), 1745-1752.
648 61. Karavalakis, G.; Bakeas, E.; Stournas, S., Influence of Oxidized Biodiesel Blends on
649 Regulated and Unregulated Emissions from a Diesel Passenger Car. Environ. Sci. Technol. 2010,
650 44, (13), 5306-5312.
651 62. U.S. Environmental Protection Agency (EPA): Toxicological Review of Hydrogen
652 Cyanide and Cyanide Salts; Washington, D.C., Septemeber 2010.
653 63. Jackson, R.; Oda, R. P.; Bhandari, R. K.; Mahon, S. B.; Brenner, M.; Rockwood, G. A.;
654 Logue, B. A., Development of a Fluorescence-Based Sensor for Rapid Diagnosis of Cyanide
655 Exposure. Anal. Chem. 2014, 86, (3), 1845-1852.
656 64. Logue, B. A.; Hinkens, D. M.; Baskin, S. I.; Rockwood, G. A., The Analysis of Cyanide
657 and its Breakdown Products in Biological Samples. Crit. Rev. Anal. Chem. 2010, 40, (2), 122658 147.
659 65. Minakata, K.; Nozawa, H.; Gonmori, K.; Yamagishi, I.; Suzuki, M.; Hasegawa, K.;
660 Watanabe, K.; Suzuki, O., Determination of cyanide in blood by electrospray ionization tandem
661 mass spectrometry after direct injection of dicyanogold. AnalBioanal Chem. 2011, 400, (7),
662 1945-1951.
663 66. Scherer, G., Carboxyhemoglobin and thiocyanate as biomarkers of exposure to carbon
664 monoxide and hydrogen cyanide in tobacco smoke. Exp. Toxicol. Pathol. 2006, 58, (2-3), 101665 124.
666 67. Rickert, W. S.; Robinson, J. C., Estimating the hazards of less hazardous cigarettes. II.
667 Study of cigarette yields of nicotine, carbon monoxide, and hydrogen cyanide in relation to
668 levels of cotinine, carboxyhemoglobin, and thiocyanate in smokers. J. Toxicol. Env. Heal. A
669 1981, 7, (3-4), 391-403.
670 68. Vinnakota, C. V.; Peetha, N. S.; Perrizo, M. G.; Ferris, D. G.; Oda, R. P.; Rockwood, G.
671 A.; Logue, B. A., Comparison of cyanide exposure markers in the biofluids of smokers and non-
672 smokers. Biomarkers 2012, 17, (7), 625-633.
673 69. Hauth, J. C.; Hauth, J.; Drawbaugh, R. B.; GILSTRAP III, L. C.; Pierson, W. P., Passive
674 smoking and thiocyanate concentrations in pregnant women and newborns. Obstet. Gynecol.
675 1984, 63, (4), 519-522.
676 70. Rickert, W.; Robinson, J.; Young, J., Estimating the hazards of "less hazardous"
677 cigarettes. I. Tar, nicotine, carbon monoxide, acrolein, hydrogen cyanide, and total aldehyde
678 deliveries of Canadian cigarettes. J. Toxicol. Env. Heal. A 1980, 6, (2), 351-365.
679 71. Graus, M.; Müller, M.; Hansel, A., High Resolution PTR-TOF: Quantification and
680 Formula Confirmation of VOC in Real Time. J. Am. Chem. Soc. Mass Spectr. 2010, 21, (6),
681 1037-1044.
682 72. Jordan, A.; Haidacher, S.; Hanel, G.; Hartungen, E.; Märk, L.; Seehauser, H.;
683 Schottkowsky, R.; Sulzer, P.; Märk, T. D., A high resolution and high sensitivity proton-transfer-
684 reaction time-of-flight mass spectrometer (PTR-TOF-MS). Int. J. Mass spectrom. 2009, 286, (2685 3), 122-128.
686 73. Ambrose, J. L.; Haase, K.; Russo, R. S.; Zhou, Y.; White, M. L.; Frinak, E. K.; Jordan,
687 C. ; Mayne, H. R. ; Talbot, R. ; Sive, B. C., A comparison of GC-FID and PTR-MS toluene
688 measurements in ambient air under conditions of enhanced monoterpene loading. Atmos. Meas.
689 Tech. 2010, 3, (4), 959-980.
690 74. Kaser, L.; Karl, T.; Schnitzhofer, R.; Graus, M.; Herdlinger-Blatt, I. S.; DiGangi, J. P.;
691 Sive, B.; Turnipseed, A.; Hornbrook, R. S.; Zheng, W.; Flocke, F. M.; Guenther, A.; Keutsch, F.
692 N.; Apel, E.; Hansel, A., Comparison of different real time VOC measurement techniques in a
693 ponderosa pine forest. Atmos. Chem. Phys. 2013, 13, (5), 2893-2906.
694 75. Kim, S.; Karl, T.; Helmig, D.; Daly, R.; Rasmussen, R.; Guenther, A., Measurement of
695 atmospheric sesquiterpenes by proton transfer reaction-mass spectrometry (PTR-MS). Atmos.
696 Meas. Tech. 2009, 2, (l), 99-112.
697 76. Tani, A.; Hayward, S.; Hewitt, C. N., Measurement of monoterpenes and related
698 compounds by proton transfer reaction-mass spectrometry (PTR-MS). Int. J. Mass spectrom.
699 2003, 223-224, 561-578.
700 77. White, M. L.; Russo, R. S.; Zhou, Y.; Ambrose, J. L.; Haase, K.; Frinak, E. K.; Varner,
701 R. K.; Wingenter, O. W.; Mao, H.; Talbot, R.; Sive, B. C., Are biogenic emissions a significant
702 source of summertime atmospheric toluene in the rural Northeastern United States? Atmos.
703 Chem. Phys. 2009, 9, (1), 81-92.
704 78. Li, R.; Warneke, C.; Graus, M.; Field, R.; Geiger, F.; Veres, P. R.; Soltis, J.; Li, S. M.;
705 Murphy, S. M.; Sweeney, C.; Pétron, G.; Roberts, J. M.; de Gouw, J., Measurements of hydrogen
706 sulfide (H2S) using PTR-MS: calibration, humidity dependence, inter-comparison and results
707 from field studies in an oil and gas production region. Atmos. Meas. Tech. 2014, 7, (10), 3597708 3610.
709 79. Vlasenko, A.; Macdonald, A. M.; Sjostedt, S. J.; Abbatt, J. P. D., Formaldehyde
710 measurements by Proton transfer reaction - Mass Spectrometry (PTR-MS): correction for
711 humidity effects. Atmos. Meas. Tech. 2010, 3, (4), 1055-1062.
712 80. Li, S. M.; Liggio, J.; Graham, L.; Lu, G.; Brook, J.; Stroud, C.; Zhang, J.; Makar, P.;
713 Moran, M. D., Condensational uptake of semivolatile organic compounds in gasoline engine
714 exhaust onto pre-existing inorganic particles. Atmos. Chem. Phys. 2011, 11, (19), 10157-10171.
715 81. Wentzell, J. J. B.; Liggio, J.; Li, S.-M.; Vlasenko, A. ; Staebler, R. ; Lu, G. ; Poitras, M. - J. ;
716 Chan, T. W.; Brook, J. R., Measurements of Gas phase Acids in Diesel Exhaust: A Relevant
717 Source of HNCO? Environ. Sci. Technol. 2013, 47, (14), 7663-7671.
718 82. Chan, T. W.; Meloche, E.; Kubsh, J.; Brezny, R., Black carbon emissions in gasoline
719 exhaust and a reduction alternative with a gasoline particulate filter. Environ. Sci. Technol. 2014,
720 48, 6027-6034.
721 83. Chan, T. W.; Meloche, E.; Kubsh, J.; Rosenblatt, D.; Brezny, R.; Rideout, G., Evaluation
722 of a gasoline particulate filter to reduce particle emissions from a gasoline direct injection
723 vehicle. SAE Int. J. Fuels Lubr. 2012, 5, (3), 1277-1290.
724 84. Makar, P. A.; Gong, W.; Hogrefe, C.; Zhang, Y.; Curci, G.; Zabkar, R.; Milbrandt, J.; Im,
725 U.; Balzarini, A.; Baró, R.; Bianconi, R.; Cheung, P.; Forkel, R.; Gravel, S.; Hirtl, M.; Honzak,
726 L.; Hou, A.; Jiménez-Guerrero, P.; Langer, M.; Moran, M. D.; Pabla, B.; Pérez, J. L.; Pirovano,
727 G.; San José, R.; Tuccella, P.; Werhahn, J.; Zhang, J.; Galmarini, S., Feedbacks between air
728 pollution and weather, part 2: Effects on chemistry. Atmos. Environ. 2014, (online 3/10/2014).
729 85. Makar, P. A.; Gong, W.; Milbrandt, J.; Hogrefe, C.; Zhang, Y.; Curci, G.; Zabkar, R.; Im,
730 U.; Balzarini, A.; Baró, R.; Bianconi, R.; Cheung, P.; Forkel, R.; Gravel, S.; Hirtl, M.; Honzak,
731 L.; Hou, A.; Jiménez-Guerrero, P.; Langer, M.; Moran, M. D.; Pabla, B.; Pérez, J. L.; Pirovano,
732 G.; San José, R.; Tuccella, P.; Werhahn, J.; Zhang, J.; Galmarini, S., Feedbacks between Air
733 Pollution and Weather, Part 1 : Effects on Weather. Atmos. Environ. 2014, (online 3/12/2014).
734 86. Yi, W.; Luan, Z.; Qiao, H.; Li, Y., Preparations of catalysts for destruction of volatile
735 organic compounds. Journal of the Chinese Rare Earth Society 2004, 22, (4), 571-574.
736 87. Ni, J.; Meunier, F. C.; Robles-Manuel, S.; Barrault, J.; Valange, S., Characterization of
737 Surface Acidity of Carbonated Materials by IR-Sensitive Molecular Probes: Advantages of
738 Using tert-Butyl Cyanide. J. Phys. Chem. C 2011, 115, (50), 24931-24936.
739 88. Thibault-Starzyk, F.; Seguin, E.; Thomas, S.; Daturi, M.; Arnolds, H.; King, D. A., Real-
740 Time Infrared Detection of Cyanide Flip on Silver-Alumina NOx Removal Catalyst. Science
741 2009, 324, (5930), 1048-1051.
742 89. Yu, Y.; Song, X.; He, H., Remarkable influence of reductant structure on the activity of
743 alumina-supported silver catalyst for the selective catalytic reduction of NOx. J. Catal. 2010,
744 271, (2), 343-350.
745 90. Yu, Y.; Zhao, J.; Yan, Y.; Han, X.; He, H., A cyclic reaction pathway triggered by
746 ammonia for the selective catalytic reduction of NOx by ethanol over Ag/Al2O3. Appl. Catal B-
747 Environ. 2013, 136-137, (0), 103-111.
748 91. Zhang, X.; He, H.; Ma, Z., Hydrogen promotes the selective catalytic reduction of NOx
749 by ethanol over Ag/Al2O3. Catal. Commun. 2007, 8, (2), 187-192.
750 92. Corrêa, S. M.; Arbilla, G., Aromatic hydrocarbons emissions in diesel and biodiesel
751 exhaust. Atmos. Environ. 2006, 40, (35), 6821-6826.
752 93. Lopes, M.; Serrano, L.; Ribeiro, I.; Cascâo, P.; Pires, N.; Rafael, S.; Tarelho, L.;
753 Monteiro, A.; Nunes, T.; Evtyugina, M.; Nielsen, O. J.; Gameiro da Silva, M.; Miranda, A. I.;
754 Borrego, C., Emissions characterization from EURO 5 diesel/biodiesel passenger car operating
755 under the new European driving cycle. Atmos. Environ. 2014, 84, (0), 339-348.
756 94. Peng, C.-Y.; Lan, C.-H.; Yang, C.-Y., Effects of biodiesel blend fuel on volatile organic
757 compound (VOC) emissions from diesel engine exhaust. Biomass Bioenerg. 2012, 36, 96-106.
758 95. Marshall, J. D.; Riley, W. J.; McKone, T. E.; Nazaroff, W. W., Intake fraction of primary
759 pollutants: motor vehicle emissions in the South Coast Air Basin. Atmos. Environ. 2003, 37,
760 (24), 3455-3468.
761 96. Holzinger, R.; Jordan, A.; Hansel, A.; Lindinger, W., Automobile Emissions of
762 Acetonitrile: Assessment of its Contribution to the Global Source. J. Atmos. Chem. 2001, 38, (2),
763 187-193.
764 97. Zhao, Y.; Strong, K.; Kondo, Y.; Koike, M.; Matsumi, Y.; Irie, H.; Rinsland, C. P.;
765 Jones, N. B.; Suzuki, K.; Nakajima, H.; Nakane, H.; Murata, I., Spectroscopic measurements of
766 tropospheric CO, C2H6, C2H2 and HCN in northern Japan. J. Geophys. Res. 2002, 107, (D18),
767 4343-.
768 98. Statistics Canada: http://www.statcan.gc.ca/tables-tableaux/sum-som/l01/cst01/trade37b-
769 eng.htm.
770 99. Karavalakis, G. ; Short, D.; Vu, D.; Villela, M. ; Russell, R. ; Jung, H. ; Asa-Awuku, A. ;
771 Durbin, T., Regulated emissions, air toxics and particle emissions from SI-DI light-duty vehicles
772 operating on different iso-butanol and ethanol blends. SAE Int. J. Fuels Lubr. 2014, 7, (1),
773 doi:10.4271/2014-01-1451.
774 100. El Ghawabi, S. H.; Gaafar, M. A.; El-Saharti, A. A.; Ahmed, S. H.; Malash, K. K.; Fares,
775 R., Chronic cyanide exposure: a clinical, radioisotope, and laboratory study. Br. J. Ind. Med.
776 1975, 32, (3), 215-219.
777 101. Standards Development Branch, Ontario Ministry of the Environment: Ontario Air
778 Standards for Hydrogen Cyanide ; 2005.
779 102. National Pollution Release Inventory carbon monoxide (CO) emissions for Canada:
780 Einvironment Canada; 2012;
781 www.ec.gc.ca/pdb/websol/emissions/ap/ap result e.cfmyear=2010&substance=co&location=C
782 A§or=&submit=Search.
783 103. Akagi, S. K.; Yokelson, R. J.; Wiedinmyer, C.; Alvarado, M. J.; Reid, J. S.; Karl, T.;
784 Crounse, J. D.; Wennberg, P. O., Emission factors for open and domestic biomass burning for
785 use in atmospheric models. Atmos. Chem. Phys. 2011, 11, (9), 4039-4072.
786 104. Global Fire Emission Database (GFED): http://www. globalfiredata.org.
791 List of Tables
792 Table 1: Averages of the Fuel-based emissions factors (aEFs) for benzene, toluene, xylenes,
793 acetonitrile and HCN from gasoline (PFI+GDI) vehicles for different transient cycles at different
794 engine temperatures at the startup.
796 Table 2: Comparison of literature HCN emission factors (mg/km) from the exhaust of different
797 light duty vehicles (LDVs)
798 Table 3: Summary of HCN, benzene, toluene, and acetonitrile EFs from ULSD and pure
799 biodiesels (B100) at different driving modes.
801 Table 4: Annual estimated HCN emissions from the transportation sector and biomass burning
802 for Ontario and Canada.
805 Table 1: Averages of the Fuel-based emissions factors (aEFs) for benzene, toluene, xylenes, acetonitrile and HCN from gasoline
806 (PFI+GDI) vehicles for different transient cycles at different engine temperatures at the startup.
Emission Factor (mg/kg of Fuel)
Vehicle Type Fuel Injection Exhaust Treatment Driving Mode Engine Temperature CO (g/km) co2 (g/km) Benzene Toluene bxylenes CH3CN HCN HCN (mg/km)
Sonata GDI GPF LA4 Cold 0.4 233.9 383.3 527.6 177.7 7.4 8.5 0.6
Sonata GDI GPF LA4 Hot 0.2 212.8 49.7 143.2 17.1 5.1 4.8 0.3
Sonata GDI GPF SS mph50 Warm 0.1 106.0 77.3 262.8 42.3 6.0 5.8 0.2
Sonata GDI GPF US06 Cold 0.8 179.0 344.9 596.6 150.6 10.8 14.9 0.8
Sonata GDI GPF US06 Hot 0.8 179.0 184.8 296.4 66.0 5.4 9.2 0.5
Volvo PFI No GPF LA4 Cold 0.2 259.5 395.7 795.3 319.6 23.4 7.9 0.6
Volvo PFI No GPF LA4 Hot 0.2 234.0 62.2 204.4 28.4 6.5 0.3 0.0
Volvo PFI No GPF S S mph50 Warm 0.1 133.1 73.9 151.0 33.4 9.0 4.7 0.2
Volvo PFI No GPF US06 Cold 0.3 217.2 91.2 217.6 63.2 5.3 16.0 1.0
Volvo PFI No GPF US06 Hot 0.3 217.2 52.9 141.1 26.9 2.1 6.0 0.4
Sonata GDI No GPF LA4 Cold 0.8 244.2 406.7 705.2 213.6 1.8 76.0 5.6
Sonata GDI No GPF LA4 Hot 0.4 217.7 103.7 321.3 48.2 1.3 46.4 3.1
Sonata GDI No GPF SS mph50 Warm 0.1 108.6 57.4 138.9 15.0 0.6 10.7 0.4
Sonata GDI No GPF US06 Cold 1.6 205.7 298.1 410.3 122.3 1.3 77.4 4.8
Sonata GDI No GPF US06 Hot 1.6 205.7 117.6 191.3 41.5 1.3 39.7 2.5
Silverado PFI No GPF LA4 cold 3.0 381.6 691.0 643.7 726.5 27.7 15.9 1.8
Silverado PFI No GPF LA4 hot 3.0 381.6 465.7 382.9 370.2 15.2 6.9 0.8
808 ^average of 22 min transient driving cycle; ^sum of o,m,p-xylene and ethyl benzene
811 Table 2: Comparison of literature HCN emission factors (mg/km) from the exhaust of different
812 light duty vehicles (LDVs)a
Min Max Average ± (1 s) Number of vehicles Model year
Keirns et al. 1978 0.8 11.8 4.5 ± 3.2 1 1977
Bradow et al. 1977 0.1 75.6 20.0 ± 24.8 49 1976
Karlsson et al. 2004 0.0 11.7 2.2 ± 4.2 5.0 1994-1998
Harvey et al, 1983 1.0 12.1 3.8 ± 4.2 206 NA
Urban et al, 1979 1.0 7.0 4.1 ± 2.6 51 1978-1979
Cadle et al, 1979 0.6 8.1 4.7 ± 3.4 26 1967-1978
This study 0.0 5.6 1.4 ± 1.7 3 2008-2011
813 a The table does not contain emission factors for idle mode.
816 Table 3: Summary of HCN, benzene, toluene, and acetonitrile EFs from ULSD and pure
817 biodiesels (B100) at different driving modes.
Emission Factor (mg/kg of Fuel)
Fuel Type Driving Mode HCN Benzene Toluene Xylene CH3CN
ULSD Idle 10.3 40 .8 85 .1 80.5 2. 1
ULSD HWFET 3.0 1.2 2.8 2.6 0.9
ULSD FTP75 3.9 1.3 3.3 2.7 0.7
ULSD US06 3.4 0.8 1.9 1.6 0.3
Canola Idle 2. 2 14.7 6. 5 3. 9 0. 1
Canola HWFET 1.2 0.5 1.2 0.7 0.3
Canola FTP75 1.2 1.3 4.0 3.0 0.4
Canola US06 1.1 0.5 2.3 0.8 0.3
Soy HWFET 0.9 0.9 3.2 1.7 0.2
Soy FTP75 0.9 1.2 2.7 1.6 0.3
Soy US06 0.8 0.9 2.2 1.2 0.2
Tallow Idle 2.0 29.5 8.7 4.1 1.3
Tallow HWFET >12 0.8 2.8 1.3 0.3
Tallow FTP75 1.1 1.1 4.1 2.0 0.2
Tallow US06 1.2 0.6 2.1 1.0 0.3
818 819
821 Table 4: Annual HCN emissions from the transportation sector and biomass burning for Ontario
822 and Canada.
Gasoline LDV (Tonnes)
Diesel LDV (Tonnes)
Total on-road vehicles (Tonnes)
Biomass Burning (Tonnes)
Ontario
Canada
2,101^-10,419^
45,496^-225,600®
824 a Estimates of annual emissions of HCN for Ontario and Canada-wide using the fuel based emission
825 factors obtained from the current laboratory study and from the 2012 annual sales for gasoline and diesel
826 for LDV reported by Statistics Canada98. 17,435,813 and 5,043,892 thousand liters of diesel were sold in
827 Canada and Ontario, respectively. On the other hand, 40,444,101 and 17,435,813 thousand liters of
828 gasoline were sold in Canada and Ontario, respectively.98 b Derived by adding HCN emissions from
829 gasoline and Diesel LDV for Ontario. c Derived by adding HCN emissions from gasoline and Diesel
830 vehicles for Canada d Value derived from using 0.00242 as HCN/CO ER22 and CO data from NPRI.102
831 'Value derived from using 0.012 as HCN/CO ER103 and CO data from NPRI.102 'Value derived from
832 using 0.00242 as HCN/CO ER22 and CO data from GFED.104 gValue derived from using 0.012 as
833 HCN/CO ER103 and CO data from GFED.104
837 FIGURE CAPTIONS
839 Figure 1: Transient exhaust levels of Toluene, benzene and HCN from a GDI vehicle not
840 equipped with a GPF as a function of different driving modes and test cycles. The PTR-ToF-MS
841 data were collected as 60 s data and the concentration in the exhaust were corrected for CVS
842 dilution.
843 Figure 2: Fuel - based emission factors (EFs) for HCN from: A) GDI vehicle (in the presence
844 and absence of a gasoline particulate filter (GPF)) and a PFI vehicle under different driving
845 modes. B) Different operation modes utilizing ULSD and B05 biodiesel blends and C) pure
846 biodiesels B100 (Tallow, Canola, Soy). All EFs calculations accounted for dilution in the CVS
847 Figure 3: Diurnal profiles for acetonitrile (ACN, green), benzene (red), toluene (blue) and HCN
848 (black) from ambient measurements in Toronto, Canada. Solid lines are means and shaded
849 regions (upper and lower bounds) indicate the 75th and 25th percentiles, respectively.
850 Figure 4: GEM-MACH Air quality model predictions for HCN over Toronto for: A) the
851 morning period (7:00- 10:00), B) evening period (16:00 - 20:00) and C) diurnal profile
852 predictions for HCN in downtown Toronto and major highway from September 7 to September
853 17, 2014.
SS mph50
Cold Hot start start
JillUUc.
■I I | I I I I I I I I I I I | I I I II I I I I I I | I I I I I I I I I M | I I I I I I" I I I I | I I I I I I I I I I I | I I I
9:00 AM 10:00 AM 11:00 AM 12:00 PM 1:00 PM 2:00 PM
Driving Time (hh:mm)
Figure 1: Transient exhaust levels of Toluene, benzene and HCN from a GDI vehicle not equipped with a GPF as a function of different driving modes and test cycles. The PTR-ToF-MS data were collected as 60 s data and the concentration in the exhaust were corrected for CVS dilution.
A) Gasoline
0 3 LL
■ LA4 Cold TU LA4 Hot
■ SS mph50 Warm
■ US06 Cold US06 Hot
B) Biodiesel (B05)
2 10-O)
ULSD SOY
CAÑOLA TALLOW
flqoJ íln,nn
HWFET FTP75 US06 Idle
C) Biodiesel (B100)
10 8 6 4 2
ULSD SOY
CAÑOLA TALLOW
HWFET FTP75 US06 Idle
865 Figure 2: Fuel - based emission factors (EFs) for HCN from: A) GDI vehicle (in the presence and absence of a gasoline particulate
866 filter (GPF)) and a PFI vehicle under different driving modes. B) Different operation modes utilizing ULSD and B05 biodiesel blends
867 and C) pure biodiesels B100 (Tallow, Canola, Soy). All EFs calculations accounted for dilution in the CV
.Q Q. Q.
0.14 0.12 0.10 0.08 0.06
£ 600 J; 500 a) 400 § 300 ■§ 200
^6.00 .Q
6 2.00 0.00
1111 ll 11 ll 11111 ll 11 ll 11111111 ll 11 ll 11111 ll 11 ll 111111111 ll | ll 11111 ll 11 ll 111111111 ll | ll 11111 ll 11 ll 11 ll ll| 11 ll |
2 4 6 8 10 12 14 16 18 20 22 Hour of Day
Figure 3: Diurnal profiles for acetonitrile (ACN, green), benzene (red), toluene (blue) and HCN (black) from ambient measurements in Toronto, Canada. Solid lines are means and shaded regions (upper and lower bounds) indicate the 75th and 25th percentiles, respectively.
•i• : i : t : • i , , i ! i i , i : :
8:00 AM 10:00 AM12:00 PM 2:00 PM 4:00 PM 6:00 PM 8:00 PM
• • I |Hwy 401 and 4001
m • • • • •
• • • . ! •• I •ill • » • < 1 • •
! : i I I I \ , 1 rrm 11 * • i •. t •
• • • • i I • • # : i | j j 1 1 1
S:00 AM 10:00 AM12:00 PM 2:00 PM 4:00 PM 6:00 PM 8:00 PM
Local Time
875 Figure 4: GEM-MACH Air quality model predictions for HCN over Toronto for: A) the morning period (7:00- 10:00), B) evening
876 period (16:00 - 20:00) and C) diurnal profile predictions for HCN in downtown Toronto and major highway from September 7 to
877 September 17, 2014.
Emissions of Hydrogen Cyanide from On-road Gasoline and Diesel Vehicles
Highlights
Samar G. Moussa, *1 Amy Leithead,1 John Liggio, *1 Shao-Meng Li,1 Tak W. Chan,2 Jeremy J.B. Wentzell,1 Craig Stroud,1 Junhua Zhang,1 Patrick Lee,1 Gang Lu,1 Jeffery R. Brook,1 Katherine Hayden,1 and Julie Narayan1
Atmospheric Science and Technology Directorate Science and Technology Branch, Environment Canada, 4905 Dufferin Street, Toronto, Ontario, Canada M3H 5T4
Emission Research and Measurement Section, Environment Canada, 335 River Road, Ottawa, Ontario, Canada, K1A 0H3
*Author to whom correspondence should be addressed: John.Liggio@ec.gc.ca; phone (416) 7394840; FAX (416) 739-4281
*Author to whom correspondence should be addressed: Samar.Moussa@ec.gc.ca; phone (416) 739-4942; FAX (416) 739-4281
1. HCN emission factors are reported as a function of fuel types and driving modes.
2. HCN emissions using diesel and biodiesel fuels are 10 times less than gasoline.
3. Ambient measurements of HCN were conducted in traffic impacted areas in Toronto.
4. Model predictions and ambient measurements of HCN were consistent with each other.
5. Exposure to HCN in urban areas is dominated by vehicle emissions.
