Influences of Waste Cooking Oil-Based Biodiesel Blends on PAH and PCDD / F Emissions from Diesel Engines in Durability Testing Cycle

In this study, the 60,000-km durability tests were performed on two diesel engines (EURO IV and EURO II) by using B10 (10% waste cooking oil + 90% diesel) and B8 (8% waste cooking oil + 92% diesel), respectively, to determine the impacts on the emissions of polycyclic aromatic hydrocarbons (PAHs) and polychlorinated dibenzo-p-dioxins and dibenzofurans (PCCD/Fs). The above emissions were measured per 20,000-km testing intervals. The highest total PAH mass concentrations were 38.2 and 25.6 μg Nm before durability test by operating the EURO IV and II engines, respectively, and decreased 51–55% after 60,000-km operation. The dominant congeners of PAH emissions were naphthalene (> 45%), pyrene, and phenanthrene, which belonged to the LM-PAHs. The total PAH BaPeq had different emission trends between the two engines during the durability tests. The highest level was 2.17 μg BaPeq Nm from EURO II engine before the test and reduced 84% after a 60,000-km cycle, when the total-BaPeq emissions of EURO IV tended to increase from 0.0894 to 0.154 μg BaPeq Nm after the same test. The most dominant congener to the toxicity emissions was benzo(a)pyrene (~70%). Additionally, the PCDD/F emissions were tested in EURO IV engine by using B10. The PCDD/F concentrations of mass and toxicity approached the highest levels, 167 ng Nm and 3.69 pg WHO-TEQ Nm, after 60,000-km and 20,000-km running cycles, respectively. The main dominant congeners were OCDD (> 50%) for mass, 2,3,7,8-TeCDD (> 35%) and 1,2,3,7,8PeCDD (> 18%) for toxicity. Consequently, the use of WCO-biodiesel might reduce the PAH mass and toxicity emissions in older engine but had no significant effect in PAH and PCDD/F emissions during the deterioration of a newer engine.


Diesel Engines and Their Emissions
As the fossil fuel crisis is getting more and more global attention, the demand for alternative fuel or green energy is becoming even stronger than before.Diesel engines were reported as one major mobile emission sources of pollutants, including carbon monoxide (CO), total hydrocarbons (HC), nitrogen oxides (NO x ), and particulate matter (PM).
PAHs and PCDD/Fs are semi-volatile species and classified as persistent organic pollutants (POPs).They are recognized as two of the potentially toxic pollutants emitted from heavyduty diesel vehicles to the atmosphere (Chang et al., 2014a;Mwangi et al., 2015b).Considering that PAH compounds and PCDD/Fs are known human carcinogens, the carcinogenic potencies of PAH emissions and PCDD/Fs especially from anthropogenic mobile emission sources, including diesel engine have drawn considerable attention for researchers (Cheruiyot et al., 2016;Chen et al., 2017).Diesel engines have higher PM emissions, which provide large surface area and carbon sources to adsorb and generate the organic compounds and lead to PAH emissions (Wichmann, 2007;Laroo et al., 2011;Cheruiyot et al., 2015).Additionally, the PCDD/F emission inventories of various regions indicated that the diesel heavy-duty trucks became one of the dominant sources, which poses a key concern due to its proximity to residential areas (Cheruiyot et al., 2016).

Biodiesel Development, Application, and Their Emissions
There are numerous studies focusing on the biodiesel as one of the most promising alternatives of diesel.They were produced from animal fats or vegetable oils by transesterification reaction and could mitigate the increase in greenhouse gas emissions (Yang et al., 2007;Lee et al., 2011;Williams et al., 2011;Mwangi et al., 2015a;Scheepers, 2015).Several advantages by using biodiesels have been reported.The higher oxygen content of biodiesels than petroleum diesel can produce more complete combustion and lower the emissions of CO, PM, HC and persistent organic pollutants (Raheman and Ghadge, 2007;Özener et al., 2014).Additionally, the lower sulfur and almost no aromatic compositions could reduce the SO 2 and POP emissions from the diesel engine and further reduce the potential of secondary fine particle formation (Tsai et al., 2015).The safety of storage and transport could also be enhanced by the higher flash point of biodiesel than that of fossil fuel (Li et al., 2016;de Mello et al., 2017).
However, the food demand, land use change, excessive water and energy demand, and secondary waste water treatment became the serious challenges of the first-generation of biodiesel produced from energy crops (Demirbas, 2008;Pimentel et al., 2009;Lin et al., 2011).In densely populated countries, such as Taiwan, the waste-cooking oil (WCO) is considered as the most promising feedstocks of biodiesel production, which does not need farming and huge extra water supply, as well as recycling the waste from human daily activity.However, WCO contain five times higher chlorine content than those in the conventional diesel due to the sauces used during food preparation (Chang et al., 2014a;Tsai et al., 2016).The higher chlorine might be responsible for causing more PCDD/F emissions from WCO combustion.Fortunately, Chang et al. (2014) conducted the research of WCO-based biodiesel and found the lower POP (PAHs, PCDD/Fs, PCBs, PBDEs, PBDD/Fs) emissions.The more complete combustion of WCO spray with higher oxygen and lower aromatic contents were the reasons of POP reduction (Chang et al., 2014a).
During application, the viscosity of biodiesel is so high it generates some problems.For example, deposits and obstruction, injector chocking, piston ring sticking and poor atomization are very common when using biodiesel (Ramadhas et al., 2005).Moreover, higher viscosity may have bad influence on spray properties and cause incomplete combustion (Muralidharan and Vasudevan, 2011).Therefore, it is like a trade-off problem between higher oxygen content and higher viscosity.It should be investigated which property is dominant in what specific situation by durability tests.

Durability Test and Deterioration of Diesel Engine
The emissions from diesel engines depend on several factors such as fuel content, engine type, loads, speed, mileage, availability of catalysts and operational conditions like temperature, pressure, injection timing, humidity, and exhaust gas recirculation (Cheruiyot et al., 2015).In the aspect of emission control, the deterioration of diesel engine is the most comprehensive analysis to express the fuel suitability for a real-scale engine.The engine durability tests then took place for testing petroleum diesel and biodiesel in the previous studies.
Engine wear and friction are the major problems after a long-term operation.The wear usually occurs on the sliding units, including cylinder liner, cam, bearing, tappet, crankshaft, pistons and pins, valves, and guides…etc.The fuel itself acted as the lubricant for the above components, when the biodiesels were reported to have better lubricity than the petroleum diesel.There were oxygen containing trace compounds, such as esters and fatty acids adsorb on rubbing surface to reduce the adhesion and limit the friction and wear formation (Haseeb et al., 2010).Many studies focused on the engine wear by long-term operation with static engine or field trail tests.The biodiesel blends (B20 to B100) were reported to have similar or even 30% less wear as compared to petroleum diesel after the durability tests (2-4 years, 200-500 h, or 30,000-200,000 km), because the unsaturated molecule could have better lubricity than the saturated ones (Clark et al., 1984;Kalam and Masjuki, 2002;Agarwai et al., 2003;Fraer et al., 2005;Proc et al., 2006;Agarwal et al., 2008).On the other hand, the biodiesels might lose their proven lubricity by auto-oxidation and corrosiveness for material degradation (Fazal et al., 2011).Therefore, the lubricity of biodiesel blends could be the competitive results by the above opposite effects.For the deterioration in diesel engine emission, Hsieh et al. (2011) reported higher PCDD/F concentrations with greater mileage of diesel engine.Yang et al. (2007) also reported that total PAH concentrations and their BaP eq emissions could be reduced by using B20, when HC, CO, PM, and particulate phase PAH emissions increased with accumulated mileage.The higher viscosity could be the reason that led to several durability issues, such as deposition formation, carbonization of injector tip, fuel filter plugging (Saxena and Maurya, 2016) and cause the increasing emissions.
Despite a good number of researches on biodiesel as an alternative fuel that has been widely investigated in diesel engines, there have been limited studies on PAH and PCDD/F emissions comparing accumulated mileage test involving different generations of engines, especially for the use of WCO-based biodiesel blends.Therefore, in this present study we will examine the PAH and PCDD/F emissions from a diesel engine by a series of durability tests.The objective of the study was to determine the longterm effects in terms of engine performances and emissions by using WCO-based biodiesel blends.

Fuel Property
Two WCO-based biodiesel blends were tested in this study: B8 (8% WCO-based biodiesel + 92% pure petroleum diesel) and B10 (10% WCO-based biodiesel + 90% pure petroleum diesel).The test fuels met the requirements of CNS 1471 (Chinese National Standards, CNS), provided by CPC corporation, Taiwan.The fuel properties were provided in Table 2. B8 was tested in Mitsubishi 4M40-0A (EURO II) while B10 was used in Mitsubishi 4M42-4AT2 (EURO IV).Based on the fuel important properties considered in the test, B8 has Cetane Index of 54 and viscosity (40°C) of 3.12 cSt, while B10 has 50.2 and 3.15 cSt for Cetane Index and viscosity (40°C), respectively.For the aspect of energy density, both two fuel blends had only < 1% lower heating values (42.76 MJ kg -1 for B8 and 42.64 MJ kg -1 for B10) than that of the petroleum diesel (42.98 MJ kg -1 ).The oxygen contents of B8 and B10 were 1.03% and 1.17%, respectively, much higher than pure diesel and potentially provide more complete combustion.

Engine, Durability Test and Sampling Procedure
The Mitsubishi 4M42-4AT2 (EURO IV) engine and Mitsubishi 4M40-0A (EURO II) engine were used in this study.The specifications of two test engines were shown in Table 1.In durability test, the test procedure divided into four steps: start, warm-up, accumulation and shutdown.The engine was started with the speed of 800 rpm for 4 min, then the engine's speed was tuned up from 1,500 to 2,000 during a 9-min warm up.Subsequently, the engine was continuously run at 3700 rpm for 13 h per day and accumulated the test hours (or mileages).The sampling was conducted at 0, 125, 250, 375, and 500 hours.The accumulation duration can be converted to real life mileage accumulation and replicate onroad conditions based on fuel consumption rate.The accumulation durations of 0, 125, 250, 375, and 500 hr are approximated to 0, 20,000, 40,000, 60,000, and 80,000 km respectively.The detail was provided in our previous study (Yang et al., 2007).
At every selected accumulated mileage, the exhaust gas samples were collected from the test engines by an isokinetic sampling system, which was specified by the U.S. EPA, modified Method 5 (USEPA, 2001) for PAH and PCDD/F analyses.A heated sampling probe (120 ± 14°C) was employed to collect both PAH and PCDD/F samples simultaneously.The target pollutants in particulate phase were collected with a glass fiber filter.Subsequently, a cooler was equipped to lower the exhaust temperature to less than 5°C before a two-stage glass cartridge set.The gaseous POPs were then collected by the cartridges.Specifically, each cartridge was packed with a 5.0 cm thickness (approximately, 20 g) of XAD-2 resin sandwiched between two 2.5 cm-thick polyurethane foam plugs.The same procedure was also applied in the durability test.

Analytical Processes
The PAH and PCDD/F samples were analysed in accordance with the US EPA modified Method 23 and were carried out in the Super Micro Mass Research and Technology Center in Cheng Shiu University, Taiwan.

PAH Analysis
The samples for PAH were extracted with a Soxhlet extractor using a mixture of n-hexane and dichloromethane (v:v = 1:1; each 250 mL) for 24 hours.The extracts were then concentrated by gently purging a stream of ultra-pure Nitrogen followed by passing through a silica gel column to be cleaned.The effluents were then re-concentrated to precisely 1 mL and moved to vial.There were 16 USEPA priority PAHs detected with gas chromatography/mass spectrometry (GC/MS) (HP 5890A/5972) equipped with a capillary column (HP Ultra 2, 50 m × 0.32 mm × 0.17 µm).

PCDD/F Analysis
After the aforementioned PAH analysis, the sample solutions were subsequently analyzed for 17 PCDD/Fs.The solutions were pre-treated with the concentrated sulfuric acid, followed by a series of sample clean-up procedures, including a multi-layered silica and alumina columns.Finally, the activated carbon columns were eluted by 40 mL toluene before instrumental analysis.A high-resolution gas chromatograph (HRGC), combined with a high-resolution mass spectrometer (HRMS), was used to measure PCDD/Fs.The HRGC (Hewlett Packard 6970 Series) is gas chromatograph equipped with a DB-5 (J&W Scientific, CA, USA) fused silica capillary column (60 m, 0.25 mm ID, 0.25 mm film thickness), splitless injection, and an initial oven temperature of 150°C.The HRMS (Micromass Autospec Ultima, UK) is a mass spectrometer with a positive electron impact (EI+) source using the selected ion monitoring (SIM) as the analyzer mode operating at a resolving power of 10000.The electron energy was fixed at 35 eV, and the operational temperature was set to 250°C.The operating details of HRGC/HRMS and the procedure for measuring of PCDD/Fs are given in our previous works (Wang et al., 2003a, b;Wang et al., 2010).

Quality Assurance and Quality Control (QA/QC)
Before the sampling, the glass fiber filters were placed in an oven at 450°C for 8 h to remove all organic compounds.The cleaned filters were kept in a desiccator for at least 8 h to maintain their fixed humidity levels prior and after the sampling.For PAHs, a breakthrough test was conducted with a three-stage glass cartridge in preliminary sampling work.The results showed the mass of 16 individual PAHs in the third stage was only 0.582-4.07%that in the total three stages.The two-stage glass cartridges deployed in this study should thus ensure a 95% collection efficiency.For PCDD/Fs, the respective surrogate standards (13Clabeled PCDD/Fs) were spiked into the cartridges and their recoveries were used to analyze the breakthroughs of POPs in the study samples.Before each diesel engine test, a leak check was carried out between the inlet of the filter holder and the outlet of the flow meter (Chang et al., 2014b).For all compound analyses, they went through the pretreatment procedures where both the internal and recovery standards were added to the sample.All procedures including the recovery of precision and recovery (PAR), surrogate, and internal labeled standards for both PAH and PCDD/Fs all complied with the relevant standards.

Emission Differences of 2 Test Engines
All experimental conditions for EURO IV and II engines as well as for B10 and B8 in this study were controlled.It is important to compare the emission differences of the two engines, because the results can be used to check whether the differences of emission results are attributed to the fuel or to engine differences.In this study, the two engines were fuelled with B10 and B8, respectively, when the emissions of PAHs and PCDD/Fs were compared to verify the emission differences of the two engines at the beginning (0 km) and the end (80,000 km for only EURO IV for PAHs) and (60,000 km for EURO II for both PAHs and PCDD/Fs) of mileage accumulation.

PAHs Emission from Diesel Engines Euro IV Engine
The concentration of 16 PAHs emitted from Mitsubishi 4M42-4AT2 (EURO IV) engine fuelled with B10 were listed in Tables 3 and 5, respectively.For EURO IV, the total PAHs concentrations were 38.2, 33.2, 26.2, 18.8 and 14.4 µg Nm -3 for 0, 20,000, 40,000, 60,000, and 80,000 km.The highest concentration of 38.2 µg Nm -3 was emitted at 0 km and then decreased to the lowest of 14.4 µg Nm -3 at 80,000 km, representing a reduction of 62%.Both the LM-PAH and MM-PAH concentrations had a reduction of 64 and 44%, respectively, whereas HM-PAHs have an increase of 113% at 80,000 km, compared to 0 km.The respective PAH congener profile fraction was shown in Fig. 1.The profile show Naphthalene as the most dominant congener (> 75%) followed by Pyrene, and Phenanthrene (each contributed > 5%); the 3 congeners all belong to the LM-PAHs.
The present study applies toxic equivalency factors (TEFs) for the 16 selected PAHs to assess the carcinogenic potencies for B10 and B8.The TEFs are useful for the comparison of the carcinogenic potency of individual PAHs in reference to the carcinogenicity of BaP (Nisbet and LaGoy, 1992).Consequently, the total BaP eq represents the carcinogenic potency of the total PAH mass concentrations (Yang et al., 2017).The BaP eq of 16 PAHs emitted from Mitsubishi 4M42-4AT2 (EURO IV) engine fuelled with B10 were listed in Table 3.The total BaP eq were 0.0894, 0.104, 0.125, 0.154, and 0.161 µg BaP eq Nm -3 for 0, 20,000, 40,000, 60,000, and 80,000 km, respectively.The lowest PAH BaP eq of 0.0894 µg BaP eq Nm -3 was emitted at 0 km, then it continued to increasing until 80,000 km, exhibiting an increase of 80%.The BaP eq for LM-PAHs showed a decrease of 52%,  whereas MM-PAHs and HM-PAHs had 18% and 192% increase, respectively, at 80,000 km compared to 0 km.Benzo(a)pyrene was the dominant congener contributing between 28.3% (0 km) to ~71.0% (80,000 km) to the toxicity followed by Naphthalene ranged between ~38.3% (0 km) to 7.2% (80,000 km), Benzo(a)anthracene, and Dibenzo(a,h)anthracene.All the 4 dominant congeners belong to HM-PAHs except Naphthalene which is a LM-PAH.
The above result showed the PAH emissions of mass could be inhibited by B10 even in the long-term durability test, since the most dominant Naphthalene could be reduced by more complete combustion provided by fuel oxygen and higher Cetane Index than petroleum diesel.However, the higher-ring PAHs, which have higher toxicity, were not obviously decreased, because the HM-PAH emissions of petroleum diesel were originally low, and could not be strongly reduced by enhancing the combustion.
Our results of PAH and PCDD/F emissions from EURO IV engine fuelled with B10 were supported by the previous study (Chang et al., 2014a).

Euro II Engine
With regards to EURO II, the respective levels of PAHs emitted at 0, 20,000, 40,000, and 60,000 km were 25.6, 19.7, 17.2, and 11.4 µg Nm -3 as shown in Table 5.The highest level of 25.6 µg Nm -3 was emitted at 0 km and then dropped until it reached the lowest level of 11.4 µg Nm -3 at 60,000 km; a reduction of 55%.The levels for LM-, MM-, and HM-PAHs all showed a similar pattern; a decrease of 51, 62, and 51% at 60,000 km, respectively, compared to 0 km.The respective PAH congener profile fraction was shown in Fig. 3.The dominant PAH species were Naphthalene contributing more than 45%, followed by Pyrene, Fluoranthene, and Phenathrene.The dominant congeners all belong to the LM-PAHs except Fluoranthene, a MM-PAH.This typical pattern was also reported in a study that analyzed the relative risk associated with PAH emissions from diesel exhaust and highlighting the importance of naphthalene as a dominant specie (Shah et al., 2005).This was concluded based on the level of the emission rate of naphthalene in comparison to other congeners.Lee et al. (2011) also reported LM-PAHs as dominant congeners of total-PAH masses.Mwangi et al. (2015) investigated PAH emissions from a diesel engine using base diesel and fuel blends and observed a notable drop of 22.0-59.5% in PAH emissions when fuel blends were used.Yang et al. (2007) used a fuel blend of B20 (80% diesel + 20% methyl ester derived from waste cooking oil) to operate Mitsubishi (4M40-2AT1) diesel engine, and they observed a PAH emission reduction of 23.7%.
Biodiesel is known to have higher oxygen content and due to the increase in methanol, it may improve the combustion performance and promote the breakdown of benzene hence reducing benzene emissions (Cheung et al., 2009).Chang et al. (2014a) suggested that the use of WCObased was responsible for reducing the precursors of the PAHs and enhanced more complete combustion in the diesel engine.In another study, Lee et al. (2011) observed a negative correlation between the biodiesel blends and the total PAHs emissions and total BaP eq .They suggested several factors causing this phenomenon including the enhanced combustion efficiency, improved PAH oxidization of forming precursors due to increased oxygenation, dilution effect which reduces fuel PAH content, and absence of aromatic content in the biodiesel (Chang et al., 2014a).Furthermore, studies have reported that fuel at optimum conditions enhances chemical reactions of unsaturated hydrocarbons, which are precursors responsible for PAH emissions that may reduce total PAH emissions (Inal and Senkan, 2002;Cheruiyot et al., 2015).
The PAH emissions from EURO II engine fuelled with B8 had a different trend in BaP eq concentration during durability test.The reductions by using WCO-based biodiesel blends in all PAH species were found more significantly in the engine with higher mileage.This might be resulted from not only the inhibition of PAH formation during engine combustion, but also have cleaning effect to the wear and friction area to reduce the carbonates and particle on the surface of engine units and further reduce the MM-and HM-PAHs.Therefore, the WCO-based biodiesel blends were observed to reduce the PAH emissions more effectively in the older engine model.

CONCLUSION
The 80,000-and 60,000-km durability tests for EURO IV and II engines by using 10% and 8% additions of WCO, respectively, in diesel fuels, showed varied results on PAH and PCDD/F concentration and toxicity emissions.The highest total PAH mass concentrations were found to be 38.2 and 25.6 µg Nm -3 before the test by using EURO IV and II engines, respectively, and decreased 62 and 55% after 80,000-and 60,000-km operation, respectively.The dominant congeners of PAH emissions were naphthalene (> 45%), pyrene, and phenanthrene, which belong to the LM-PAHs.The total PAH BaP eq had different emission trends between the two engines during the durability tests.The highest level was 2.17 µg BaP eq Nm -3 from EURO II engine before test and reduced 84% after a 60,000-km cycle, when the total-BaP eq emissions of EURO IV tended to increase from 0.0894 to 0.161 µg BaP eq Nm -3 (80% increase) after the 80,000-km testing cycle.The most dominant congener contributing to the toxicity emissions of engines was benzo(a)pyrene (~70%), which had higher molecular weight and tend to be in particulate phase (Lee et al., 2011).Additionally, the PCDD/F emissions were tested in EURO IV engine by using B10.The PCDD/F concentrations of mass and toxicity approached the highest levels, 167 ng Nm -3 and 3.69 pg WHO-TEQ Nm -3 , after 60,000-km and 20,000-km running cycles, respectively.The main dominant congeners were OCDD (> 50%) for mass, 2,3,7,8-TeCDD (> 35%) and 1,2,3,7,8-PeCDD (> 18%) for toxicity.Consequently, the use of WCO-biodiesel might reduce the PAH mass and toxicity emissions in older engine (EURO II) but showed no significant effect in PAH and PCDD/F emission during the deterioration of a newer engine (EURO IV).

Fig. 1 .
Fig. 1.Mass fractions of PAH congeners in the emission of EURO IV engine by using B10.

Fig. 2 .
Fig. 2. Mass fractions of PCDD/F congeners in the emission of EURO IV engine by using B10.

Fig. 3 .
Fig. 3. Mass fractions of PAH congeners in the emission of EURO II engine by using B8.

Table 2 .
Specifications of tested diesel engine.

Table 3 .
PAH mass and BaP eq concentrations emitted from EURO IV diesel engine fueled with B10.

Table 4 .
PCDD/F concentrations and toxicities emitted from EURO IV diesel engine fueled with B10.

Table 5 .
PAH mass and BaP eq concentrations emitted from EURO II diesel engine fueled with B8.