Influence of Biodiesel on Emissions of Criteria and Non-Criteria Primary and Secondary Pollutants

Abstract

Considering the plan to increase the mixture of biodiesel in diesel oil (currently 14% in Brazil), it is becoming necessary to know the composition of exhaust gases. This study shows that most recent articles indicate that biodiesel is a suitable alternative in these circumstances, although they point out that it can have harmful effects, such as an increase in some emissions, for example carbonyls. A diesel utility van was used fueled by different mixtures of 0, 10, 15, 20 and 30% of biodiesel added to diesel, under a protocol test using a dynamometer. The emissions were collected from a constant volume sampler and transferred to a 4 m³ polytetrafluoroethylene (PTFE) reaction chamber to enable the formation of secondary pollutants, which allowed the study to be carried out without the influence of meteorological parameters. The results are highlighted for each of the pollutants and show the differences in the composition of the exhaust gases when the fuel is changed. No significant changes were observed in carbonyls emissions, but a clear reduction for benzene, toluene, ethylbenzene and xylenes (BTEX) and alkyl-PAHs (polycyclic aromatic hydrocarbons), and an increase in nitro-HPA were detected. Fine particles (< 1.8 µm) also showed an increase, but nano particles (< 180 nm) demonstrated a decrease.

Keywords:
emissions; PAHs; BTEX; carbonyls; biodiesel; chamber

Introduction

The air quality of large urban centers has been deteriorating as a result of increasing industrialization and urbanization. Air pollution from anthropogenic activities is one of the main causes of environmental degradation since it is harmful to human health as well as damaging fauna and flora. Among the sources of pollution, fuel combustion is one of the main contributory factors to the increase of pollutant emissions into the atmosphere.¹ It should be noted that at the beginning of 2020, the appearance of the world pandemic of COVID-19 (coronavirus disease 2019) changed the pattern of air pollution in several regions of the world. In the first region affected by the virus, the city of Wuhan in China, the authorities imposed quarantine restrictions that resulted in a reduction of up to 30% in emissions of nitrogen dioxide (NO₂) and up to 25% in carbon dioxide (CO₂).²,³ The authors also reveal that, in light of the enormous reduction in air pollution after the quarantine, the COVID-19 pandemic may, paradoxically, have reduced the total number of deaths or respiratory diseases during that period that are attributed to air pollution.⁴ A significant reduction in NO₂ and particulate matter below 2.5 µm (PM_2.5) emissions was also observed in Europe during the quarantine, such as in France, Germany, Italy and Spain³ and also in Brazil.⁵,⁶ Other authors,² however, claim that the significant reduction of emissions in transport and industry during the quarantine period in China was not enough to significantly reduce air pollution in the country, especially when meteorological conditions are unfavorable.

Since the 1940s, several vehicle emission control technologies have been developed and introduced to reduce air pollution. In Brazil, since the setting up of PROCONVE (Air Pollution Control Program for Motor Vehicles),⁷ in 1986, there has been a gradual reduction in the levels of vehicular emissions.

The chemical composition of a city’s atmosphere is a result of several factors, such as emissions from mobile sources (vehicles), stationary sources (factories, homes and landfills) and natural sources (natural fires and volcanic activities). Pollutants can be classified as primary and secondary, depending on their origin. The former are emitted directly from the sources and the main ones coming from vehicles are NOx (NO + NO₂), carbon monoxide (CO), CO₂, volatile organic compounds (VOCs), methane (CH₄) and particulate matter (PM). Secondary pollutants are the result of chemical reactions of primary pollutants in the atmosphere, which is the case of tropospheric ozone.¹

In certain atmospheric conditions, VOCs can react with NOx in the presence of sunlight, and can form ozone in the troposphere, with harmful effect on human health and air quality. The chemistry of ozone formation is not linear, which makes it more complex and difficult to control, when compared with other pollutants.⁸,⁹

The World Health Organization (WHO) published first guidelines in 2005,¹⁰ and updated in 2021,¹¹,¹² on a global scale, with the aim of limiting the maximum levels of tropospheric ozone concentrations. In its target, the maximum average concentration in 8 h of exposure per day should not exceed 100 µg m^-3 to ensure a reasonable degree of protection for human health.

Diesel is a widely known fossil fuel since it is one of the main sources of gaseous and particulate emissions in cities. Although trucks and buses represent only 5.66% of the Brazilian national fleet,¹³ diesel consumption represents 51.9% of total fuels in Brazil (140,135,096 m³ in 2022).¹⁴ The use of biodiesel as an alternative fuel is a propose to reduce the emissions of pollutants and also make drivers in the future less dependent on fossil fuels; however, the physical and chemical properties of biodiesel are not the same as those of diesel oil. Moreover, different biodiesels may differ from each other, as their composition depends on the oilseeds from which they were extracted.¹⁵,¹⁶ The National Biodiesel Production and Use Program (PNPB) was set up in 2004 by the Federal Government;¹⁷ this had a technical basis and was driven by powerful environmental, social and marketing concerns. It recommended the progressive use of biodiesel in addition to mineral diesel oil. In its initial phase, in 2005, the addition of 2% of biodiesel to diesel oil was authorized, followed by a planned increase in content, which culminated in the mandatory addition of 5% of biodiesel as of 2010. Law No. 13.263 of March 23, 2016,¹⁷ established a scheme for the gradual increase of the proportion of biodiesel to diesel oil. The percentage of biodiesel added to diesel oil increased from B7 to B8, from B8 to B10, from B10 to B11 and later to B12, in the years 2017, 2018, 2019 and 2020, respectively. However, the same law stipulates the need to carry out tests for the technical validation of the B10 mixture, within 12 months of its promulgation. The biodiesel content increased to 13% in March of 2021, 14% and in March of 2024 and 15% in August of 2025. In 2024, biodiesel represented 6.8% of total fuels produced in Brazil.¹⁸

In Brazil, the actual production of biodiesel has a contribution of 70.5% from soybean, 11.2% from animal fat, 13.2% from other fatty materials and 5.1% from other sources.¹⁸

Validation of the use of biodiesel at high levels is essential to understand the composition of vehicle exhaust when diesel-biodiesel mixtures are used, as well as the ability of vehicles to form secondary pollutants and the possible effects on public health. This knowledge makes it possible to form the composition of future mixtures, while always seeking to know the effects of each on the troposphere.¹⁷,¹⁹

Although biodiesel is an alternative means of reducing the emissions of some pollutants, its use can lead to some negative outcomes. In addition to the fact that it has a higher viscosity than diesel, which causes heterogeneity in the combustion of biodiesel, it has a lower calorific value and is more expensive; it even increases the emission of NOx, an important ozone precursor and also carbonyls compounds.¹⁹

In a study²⁰ carried out on the bus fleet of Curitiba, it was observed that vehicles powered by B100 (pure biodiesel) had an average emission that was 26% higher than that of vehicles fueled with diesel S10, when compared with vehicles with the same type of engine. The results indicated an increase in NO₂ emissions and, in particular, NO when engines were fueled with biodiesel.

The aim of this study is the measurement of criteria primary pollutants (CO, CO₂, HC, NOx and PM) from the exhaust gases of a light commercial vehicle fueled with different levels of biodiesel (B0, B10, B15, B20 and B30), and the formation of primary and secondary pollutants as carbonyls, BTEX (benzene, toluene, ethyl benzene and xylenes), polycyclic aromatic hydrocarbons (PAHs) and its derivatives.

Experimental

Vehicle emissions research laboratory: primary pollutants

A light commercial vehicle was used for the determination of hydrocarbons (HC), CO, NOx, CO₂ and PM in the exhaust gas and for the determination of aldehydes by liquid chromatography, BTEX and PAHs by gas chromatography and tested on a chassis dynamometer. The engine specifications are listed in Table 1.

Table 1
Vehicle specifications used in the experiments

Model	Light commercial
Manufacturer	Renault
Displacement	2.5 L with 16 valves
Power	115 cv (84 kW) at 3500 rpm
Torque	29.6 kgf (290 Nm) at 1600 rpm
Odometer / km	71,000
Year	2012
Phase	Proconve P7

The vehicle has an after-treatment system (DOC, diesel oxidation catalyst), and an exhaust gas recirculation valve (EGR). The exhaust gases were diluted (1:20) with the dilution air in a constant volume sampler (CVS) system for later collection and determination of the chemical composition. A Horiba Mexa 7200 (CO, HC, NOx, CO₂, CH₄) bench analyzer and 7500 DEGR (CO, HC, NOx, CO₂, O₂) were used to quantify the compounds. The total hydrocarbon (THC) content was determined by flame ionization detection (HORIBA, model FIA-720, 0 50 ppmC), while CO and CO₂ content was determined by non dispersive infrared spectrometry (HORIBA, model AIA 721A, 0-200 ppm, and AIA-722, 0-2.0% v/v, respectively) and NOx by chemiluminescence (HORIBA, model CLA-720A, 0-50 ppm).

An activated charcoal cartridge (Supelco ORBO 32 400/200 mg) was used for BTEX sampling at a flow rate of 1.5 L min^-1. One cartridge was used for each phase of the standard driving protocol, and another cartridge for the collection of dilution air. Following this, the contents of each cartridge were transferred to a 2 mL vial and added to 1000 µL of dichloromethane at -20 °C to prevent the volatilization of the lighter BTEX, since the extraction process is exothermic. The flasks were capped with septum caps, placed in an ultrasonic bath for 20 min, and then allowed to rest for 1 h.

The BTEX chemical analyses were performed by gas chromatography with mass spectrometry (GC-MS) on a Varian 450GC 220MS (Walnut Creek, CA, USA) chromatograph using a VF-5MS column (30 m, 0.25 mm and 0.25 µm). Injections of 1.0 µL of a sample were conducted at 200 °C, with a split ratio of 1:4, using helium 5.0 as a carrier gas at 2.0 mL min^-1. The initial column temperature was 40 °C, which was maintained for 3 min, and then followed by a heating rate of 15 °C min^-1 up to 200 °C, which was held for 6 min. The temperatures of the ion trap, manifold and transfer line were 150, 40 and 180 °C, respectively. The MS detector monitored ions from m/z 72 to 79, 89 to 93, 101 to 107, and 119 to 121.

The calibration was performed with a standard BTEX solution (Supelco EPA TO-1 Mix 1A) by external standardization with concentrations of 0.1, 0.5, 1.0, 2.0 and 4.0 ng mL^-1, injected in triplicate, with an acceptance criterion of the analytical curve determination coefficients higher than 0.99. The calculation of the limit of detection (LOD) was done multiplying the standard deviation of blank signal by 3.3 and dividing by the slope of the analytical curve. The limit of quantification (LOQ) was the value of the LOD multiplied by 3.0. The LOQ calculated for all BTEX compound was 0.056 ng mL^-1, which corresponds to a concentration of 1.2 ng m^-3 in the gas phase. All the measurements were within the analytical curves for all the samples and no dilution was necessary.

The carbonyl emissions were sampled at 1.5 L min^-1 using impingers with 2,4-dinitro phenylhydrazine acid solution (2,4-DNPH) following the guidelines of ABNT NBR 12026.²¹ The sampling encompass all three stages of the test with a total time of 31.27 min and the extraction volume of each cartridge was 1000 µL with acetonitrile. Chemical analyses were performed by high-performance liquid chromatography on an Agilent LC1200 (Waldbronn, Germany) with a G1314D detector at 365 nm. A volume of 20 μL was injected using a ZORBAX ODS column (25 cm × 4.6 mm × 5.0 μm) maintained at 35 °C. The mobile phase employed was 65% acetonitrile and 35% water with a flow beginning at 2.00 mL min^-1 until 5 min increasing up to 3.0 mL min^-1 until the end of the chromatogram. The analytical curves were prepared by means of two types of standards, a Supelco 47650-U Mix and CRM47651 Mix, containing eight carbonyls-DNPH derivatives: formaldehyde, acetaldehyde, acrolein, acetone, propionaldehyde, butyraldehyde and benzaldehyde. Standards concentrations of 2.0, 5.0, 10.0, 20 and 40 ng mL^-1 were used, injected in triplicate, with an acceptance criterion of the analytical curve determination coefficients higher than 0.99. The LOQ was 1.16 ng mL^-1 for all carbonyls.

PAHs in the gas phase were sampled using XAD 2 cartridges (Supelco ORBO 609 400/200 mg) at 1.5 L min^-1. The heavier PAHs were determined in the polytetrafluoroethylene (PTFE) membrane (47 mm and 0.45 µm) used to clean the PM before entering the CVS. The same procedures after sampling were used for BTEX. Chemical analyses were performed using the same GC MS and chromatograph column. It was injected 2.0 µL at 320 °C in splitless mode.

The oven temperature began at 70 °C, maintained for 2 min, followed by a heating rate of 12 °C min^-1 up to 320 °C which was maintained for 8 min. The monitored ions (m/z) were 127-128 for naphthalene (NAP), 152 154 for acenaphthylene (ACY) and acenaphthene (ACE), 165 166 for fluorene (FLU), 177 178 for phenanthrene (PHE) and anthracene (ANT), 202 203 for fluoranthene (FLT) and pyrene (PYR), 225 230 for benzo[a]anthracene (BaA) and chrysene (CRY), 250 260 for benzo[b]fluoranthene (BbF), benzo[k]fluoranthene (BkF) and benzo[a]pyrene (BaP) and 272-282 for dibenz[a,h]anthracene (DBA), benzo[g,h,i]perylene (BGP), indene[1,2,3-cd]pyrene (IND). For all ions the filament current was 40 µA.

For the quantification of the PAHs, analytical calibration curves were prepared using standard solutions (Supelco EPA 610-N PAH kit 47351) at five different concentration levels from 10 to 200 µg L^-1, using the external standard technique in duplicate injections. The LOQ ranged from 2.26 µg L^-1 (DbA and BGP) to 7.22 µg L^-1 (ANT). Alkyl and nitro PAHs (APAH and NPAH) chemical analysis were performed using the same analytical range calibration and acceptance criteria, using standards from Chemservice for APAHs²² and Dr. Ehrenstorfer for NPAHs.²³

Vehicle emissions research laboratory: secondary pollutants

The gases from the CVS were transferred at the end of the tests to a chamber, built with 5 mil Teflon FEP (fluorinated ethylene propylene), inert to the exhaust gases and was permeable to UV rays. The chamber had a volume of 3.9 m³, and included an aluminum support frame, resulting in dimensions of approximately 1.19 × 1.19 × 2.78 m. Both the sides and the bottom of the chamber were covered with a reflective shield in order to increase the incident radiation, as well as to prevent people from being exposed to UV radiation. The simulation of sunlight was based on the work of Barnes and Rudzinki,²⁴ and involved using 8 fluorescent lamps (Philips TUV30W G30T8 and UV-C and Bravo F30T8/BL UV-A) that emit UV radiation with a peak at 254 and 365 nm, respectively, and placed at the top of the appliance. A 100 mm fan was placed inside the chamber (at the bottom) to achieve a better homogenization of the gases in the chamber.

The same pollutants were sampled inside the reaction chamber after 1 h of light irradiation. The determination of alkyl PAHs (APAHs) and nitro PAHs (NPAHs) were performed from the reaction chamber.

The heavier PAHs and its derivatives from the reaction chamber were sampled using a Micro Orifice Uniform Deposit Impactor (MOUDI MSP 120 R) at a flow rate of 30 L min^-1 for 2.0 h until the reaction chamber was completely empty. The ten stages of the MOUDI were grouped in particulate matter fine (PMF, < 1.8 µm), particulate matter ultrafine (PMU, < 560 nm) and particulate matter nano (PMN, < 180 nm) and were extracted using dichloromethane and concentrated to 2 mL.

Fuels and test methods

With regard to the fuels used in the tests, four mixtures of standard reference diesel fuel²⁵ with soybean biodiesel B100 were prepared in the following proportions: (i) 10% volume of biodiesel, with a density 835.0 kg m^-3 (S10-B10); (ii) 15% biodiesel, with a density of 837.5 kg m^-3 (S10-B15); (iii) 20% biodiesel, with a density of 840.0 kg m^-3 (S10-B20) and (iv) 30% biodiesel, with a density of 845.0 kg m^-3 (S10-B30). The vehicle was tested four times with each fuel to allow a statistical treatment of the data. The FTP-75 cycle (Federal Test Procedure), standardized by ABNT NRB 6601,²⁶ was adopted, and this consists of three phases: the cold start phase, the stabilized phase and the hot start phase.

The sampling of primary pollutants and their introduction into the reaction chamber took place in a mixture of air through CVS and, since it is a vehicle of the diesel cycle, the determination of the gases was carried out together with particulate material, measured by gravimetry. Figure 1 shows the layout of an emission chamber, where the vehicle has its driving training, aided by a chassis dynamometer, responsible for simulating a track load. Throughout the test, the emissions were diluted with ambient air by means of CVS, with a constant sample being removed into the bags used to integrate the emission control system of the vehicle. The same process was used to collect gases for the reaction chamber. At the end of the test, the UV lamps were turned on and the gas analyzers began to monitor the reactions inside the chamber. Four tests were performed for pure diesel, 5 tests for B10, 4 tests for B15, 3 tests for B20 and 5 tests for B30.

Figure 1
Schematic diagram of the testing procedure and reaction chamber.²⁷

Data treatment

The data analysis tools used were: Tukey test, dendrogram, correlation matrix, and barplot. The Tukey test is a parametric multiple comparison test, used to determine any contrast between two average treatment effects. Because it is both easy-to-apply and rigorous, it is one of the most widely used comparison tests.²⁸ In this work it was employed with 95% of significance. The correlation matrix is used to determine the correlations between the variables through Pearson’s coefficient, and comparing each pair of items. Pearson’s correlation coefficient (r) is the linear correlation between two variables, that is, it expresses the way two variables correlate with each other.²⁹

Results and Discussion

Primary pollutant emissions

The data in Figures 2 to 9 are displayed in the form of a barplot with standard deviations to highlight the pattern of emissions with biodiesel content.

Figure 2
Barplot for primary criteria pollutants using different biodiesel contents. Deviation bars were obtained from 4 replications.

Figure 3
Barplot for primary BTEX using different biodiesel contents. Deviation bars were obtained from 4 replications.

In the case of THC, a clear reduction was observed when the biodiesel content increased (Figure 2). It is essential to interpret the averages of the Tukey test (Figure S1, Supplementary Information) by observing the horizontal lines in relation to the zero slope of a vertical line. If the horizontal line touches the vertical, the fuel comparison between the pair is considered to be without any significant statistical difference; this is the case in the comparison between B20-B30, B15-B30, B15-B20, B10-B15 and B0-B15. When a comparison is made between B0 and B20, there are significant statistical differences, and from Figure 2 it is clear that the THC emission with B20 is lower than B0. The B0 fuel emitted an average of 0.44 g km^-1 while B30 emitted an average of 0.37 g km^-1. This tendency to reduce THC emissions has also been found in other sources such as Ghazali et al.¹⁶ and Agarwal et al.³⁰ It has also been noted that biodiesel has oxygen in its composition, which can improve combustion efficiency, by reducing THC emissions. Fossil fuels have a higher concentration of long-chain hydrocarbons in their composition than renewables ones.³¹

The non-methane hydrocarbons (NMHC) in Figure 2 follow the same pattern as THC in reducing their emissions by increasing the biodiesel content of the mixture, with the more striking differences being found between B0-B30, B10-B30 and B0-B20 (as shown in Figure S1).

With regard to the results of CO, CO₂ and NOx, no linear behavior was observed in the biodiesel content, and in the case of CO and CO₂ the equality of means/medians hypothesis test is not rejected, since neither of them differs significantly (as seen in Figure S1). In the case of NOx emissions (Figure S1) statistically significant differences were found between the mean averages of the subsequent levels, that is, B0-B20, B10-B20 and B20-B30. On the basis of a study by Man et al.³¹ and Agarwal et al.,³⁰ the authors only observed a significant influence on NOx emissions when there were mixtures with a percentage above 30%. In terms of volume, this can be quantified with CO tending to decrease due to the greater presence of oxygen in the mixture and an increase in NOx emissions.

The NOx value is very high for the tested model and the emission measurement is above the regulatory limit allowed by legislation,⁷ which for the P7 phase is 0.60 g km^-1. The production of NOx can vary depending on several factors, such as the rise in temperature during the combustion, the type and quality of the fuel and the operating and maintenance conditions of the engine.

There was a significant discrepancy in the analysis of PM emissions in Figure 2 where there was no tendency changing the biodiesel content, and the fact that there may be equality between the averages cannot rejected (as shown in Figure S1).

BTEX can be formed if there is an incomplete combustion of these compounds that are present in diesel or it might be formed during the combustion of hydrocarbons with low oxygen content. From BTEX results, a clear reduction of emissions was found when there was an increase of biodiesel content, with a sharper reduction if B15 was used rather than B20 (as observed in Figure 3), despite the high variability between the replicates.

With regard to benzene, toluene and ethylbenzene, only emissions using B0-B15, B0-B20 and B0-B30 can be considered to be statistically different (as shown in Figure S2 for the Tukey tests). However, no differences between emissions using all the biodiesel mixtures were observed for xylenes.

Amaral et al.³² reported that benzene and ethylbenzene have higher rates of emission in B100 than B5 and similar results were found in an extensive study conducted by Turrio-Baldassari et al.³³ Guarieiro et al.³⁴ found that if they used a heavy truck with an after-treatment device under a dynamometer, there was a reduction for benzene with the use of biodiesel, but no statistical significance was observed for the other mono aromatics.

Carbonyls from the engine exhaust are formed by oxidation reactions of the fuel, resulting from incomplete combustion or its by-products. Chemical reactions involving methane, twin-carbon hydrocarbons and ethanol are the main formation mechanisms of carbonyl emissions and Figure 4 shows the effects on the emissions with no clear tendence of change with the biodiesel content. As biodiesel includes oxygen in its chemical composition, this allows the oxidation mechanism to form these compounds. Other factors, such as combustion temperature, mixture efficiency and radical formation were also important, but in this work, we compare the results of same tests using different fuels mixtures. Carbonyl emissions are of great concern as they are one of the main precursors of tropospheric ozone, and have an impact on urban air quality.³⁴,³⁵

Figure 4
Barplot for primary carbonyls using different biodiesel contents. Deviation bars were obtained from 4 replications.

Figure S3 (Supplementary Information) shows that the emissions are quite similar when compared with most biodiesel blends, and the most striking differences observed for all carbonyls are those that use B0 and B30.

Karavalakis et al.³⁶ used two heavy trucks in a dynamometer test, one with after-treatment devices and another without DPF (diesel particulate filter) and DOC. They reported that formaldehyde and acetaldehyde are the dominant carbonyls, as we found in our work and corroborated by other authors.³⁷-⁴¹ Karavalakis et al.³⁶ reported a reduction of carbonyl emissions could be achieved with the use of DPF and DOC, but obtained opposite results without these after-treatment devices.

Comparing the primary PAHs emissions, where ACE, ACY and NAP were sampled in the gas phase using XAD 2 cartridges and PHE, FLT and PYR were sampled using PTFE discs, it seems that the emissions presented in Figure 5 are quite similar. The conclusion is also confirmed looking the Tukey test in Figure S4 (Supplementary Information), where only FLT emissions from B0 and B10 are different, with a confidence level of 95%.

Figure 5
Barplot for primary PAHs using different biodiesel contents. Deviation bars were obtained from four replications.

These findings are quite different from our previous works⁴²,⁴³ using the same vehicle tested without the use of a dynamometer and a standard protocol. In these previous studies it was observed a clear reduction on the PAHs emissions with the increase of the use of biodiesel. The unique difference between the tests was that the vehicle was tested under a constant rpm without load in a neutral position.

Formation of secondary pollutants

The measurement inside the reaction chamber took place after a period of one hour from the end of emissions collection, which made it possible to monitor changes in the chemical composition, and the formation or consumption processes. The results are presented in Figures 6 to 9.

Figure 6
Barplot for secondary PAHs using different biodiesel contents. Deviation bars were obtained from four replications.

It is of great importance to study the formation of secondary pollutants in the urban troposphere from the interaction of primary pollutants emitted by emitting sources under the influence of meteorological parameters. The ambient air breathed by the population of cities is the result of the mixture of primary and secondary pollutants, added to the transport of pollutants from one location to another.

From the data in Figure 6, it is possible to visualize a reduction in the formation of secondary PAHs with the increase in the use of biodiesel, especially for FLT, NAP and PYR. However, the best conclusions can be observed by the Tukey tests in Figure S5. The biggest difference observed is between B0 to B30 fuels for all PAHs, with the exception of CRY. The same observation can be made between fuels B10 to B30, to a lesser extent, and also between fuels B0 to B20 and B10 to B20.

Results for NPAH are presented in Figure 7 in the form of barplots and Figure S6 (Supplementary Information) for Tukey tests. It is possible to observe a consistent increase of the NPAH compounds with the increase of the biodiesel content in the blends. PAHs react in the gas phase with NOx, which leads to toxic compounds such as NPAHs, with carcinogenic, mutagenic and estrogenic properties, with higher risks than their PAHs analogs.⁴⁴-⁴⁷ The Tukey tests also presented the same conclusion, with pronounced differences for all blends for 9-NPHE, 9-NANT, 1-NPYR and 6-NCRY, and most of the blends for 1-NNAP, 2-NFLU and 2-NFLT.

Figure 7
Barplot for secondary NPAHs using different biodiesel contents. Deviation bars were obtained from four replications.

The APAH showed an inverse behavior of the NPAH, but similar to the PAH, but with a more pronounced tendency. Figure 8 shows a clear reduction in the formation of APAH with increasing biodiesel content in the mixture. For greater certainty on this observation, it can also be noted in Figure S7 (Supplementary Information), that Tukey’s tests also indicate a statistically significant difference for most blend pairs, except for a few blend pairs with little difference in biodiesel. The biggest differences are when using the B30 blend. There are few studies involving APAH emissions from mixtures of diesel and biodiesel and non-existent for the formation of secondary APAHs. The available studies are from Dobbins et al.⁴⁷ and California Air Resources Board (CARB).⁴⁸

Figure 8
Barplot for secondary APAHs using different biodiesel contents. Deviation bars were obtained from four replications.

Figure 9
Barplot for secondary grouped PAHs, NPAHs and APAHs separated by particle size, sampled using MOUDI, using different biodiesel contents. Deviation bars were obtained from four replications.

All PAHs in the gas phase were sampled using XAD 2 cartridges, both primary and secondary pollutants. PAHs and its derivatives were sampled in the particulate phase using the MOUDI impactor and the 10 stages were grouped in 3 ranges of size: PMF (10, 5.6, 3.2, 1.8 µm), PMU (1.0 µm, 560 nm, 320 nm) and PMN (180, 100, 56 nm).

From Figure 9 it is possible to note that the sum of all PAHs and its analogs grouped by size, show that PMN and PMU are quite similar for all biodiesel contents considering the deviation bars, with an average increase of 24.0% from to pure diesel to 30% of biodiesel. But PMF, those with higher aerodynamic diameter, shows a clear tendency of increase with the increase of biodiesel in the mixture with pure diesel, with an average value of 276% from pure diesel to 30% of biodiesel.

Multivariate statistical study

An overview of all the primary emissions was obtained by ensuring there was a linear correlation (Pearson) between the variables (Figure 10), where BTEX and carbonyls (RCHO) were grouped together for a clearer visualization. If the correlation is positive, it indicates that when one variable grows or decreases the other tends to do the same. If the correlation is negative, it indicates that when one variable grows, the other decreases and vice versa. The absolute value of the correlation represents the strength of these two variables, the higher the value (positive or negative near 1.0), the stronger the association. To check the correlation between two variables, just cross the line for the first variable and the column for the second.

Figure 10
Correlation matrix for all variables using the Pearson method.

On the basis of the correlations displayed in Figure 10, it is possible to picture a strong negative correlation between the biodiesel content and the formation of THC, BTEX, NMHC, secondary PAHs and APAHs, which suggests a greater degree of efficiency in burning the fuel. These classes of pollutants are also correlated, which suggests the same formation process is possible. Contrary to most of the literature, it was noted that there is a medium negative correlation between fuel and the formation of NOx, as well as a medium positive correlation between biodiesel and CO and CO₂. In light of this, nearly every author states that the use of fuel with oxygen in its chemical composition, leads more frequently to a conversion of CO to CO₂, although other processes that occur during the burning and even after the burning can definitely affect the CO levels. The strongest negative correlation for RCHO is with consumption. Biodiesel has a medium correlation with engine consumption (as stated earlier) owing to its low calorific value. Methane and PM have only one medium correlation with each other and NOx.

Figure 10 also provides a second classification, based on dendrograms that are calculated by means of Euclidean distances. It can be seen that biodiesel and consumption forms a group which suggests that consumption increases with biodiesel content. Another strong group is formed between NMHC + THC + BTEX as all of them are hydrocarbons, and also with secondary PAHs and APAHs. Surprisingly CH₄ is not included in this group and is positively correlated with PM and primary PAHs. RCHO + CO + CO₂ form another group and this is probably combined with same emissions and formation process.

Conclusions

This study explored the emissions from a vehicle fueled with pure diesel and with different levels of biodiesel. This vehicle is a common van used in Brazilian cities.

The NOx emissions were found to be lower with the B20 mixture; however, when compared with the maximum limits established by PROCONVE, they are almost 50% above that allowed for the tested vehicle phase.

The composition of biodiesel can cause a decrease in NOx emissions. The results obtained contradict many researches from the literature, where the addition of biodiesel leads to an increase in NOx, and in the case of some authors, this increase is more significant with mixtures above 30%.

The results for THC corroborate those found in the literature. When there is an increase in biodiesel content, there is a reduction in THC emissions. The same conclusion was reached when the NMHC results were analyzed, which suggests that there is a greater efficiency in the combustion of the engine with the use of biofuel.

However, the use of different levels of biodiesel did not have a significant impact on the average emissions of CO and CO₂, which were found to be statistically equal.

With regard to the PM, there was no statistical difference, but the average emissions were lower with mixtures of B10 and B15.

It should be pointed out that the total emission of aldehydes from the diesel vehicle is not negligible. The average was above the limit for vehicles of the Otto cycle of the same phase. Although there are no limits to aldehyde emissions for diesel vehicles, the results show that these emissions deserve further study and should not be ignored.

The best results were a reduction of BTEX and APAHs (primary and secondary), but the NPAHs showed a significant increase. For primary PAHs only FLT was different from B0 and B30. For secondary formed PAHs FLT, NAP and PYR presented a different pattern and in a lesser extent B10 from B30 and B20 and B0 from B20.

It is worth mentioning that the results obtained in this research study are restricted to a single vehicular test, so they should be regarded as partial and limited.

Supplementary Information

Supplementary Information

Supplementary information is available free of charge at http:// jbcs.sbq.org.br as PDF file.

Data Availability Statement

All data are available in the text.

Acknowledgments

The authors are grateful to the Brazilian Ministry of Science and Technology (MCT) and the Brazilian National Council for Technological and Scientific Development (CNPq), Law No. 8010/90, FAPERJ for funding this project. They would also like to thank the staff members of the LACTEC Automotive Laboratory (LEME) for their valuable assistance and support.

References

• ¹ Guimarães, C. S.; Control and Monitoring of Atmospheric Pollutants; Elsevier: Rio de Janeiro, Brazil, 2016.
• ² Wang, P.; Chen, K.; Zhu, S.; Wang, P.; Zhang, H.; Resour., Conserv. Recycl. 2020, 158, 104814. [Crossref]
  » Crossref
• ³ Zambrano-Monserrate, M. A.; Ruano, M. A.; Sanchez-Alcalde, L.; Sci. Total Environ. 2020, 728, 138813. [Crossref]
  » Crossref
• ⁴ Dutheil, F.; Baker, J. S.; Navel, V.; Environ. Pollut. 2020, 263, 114466. [Crossref]
  » Crossref
• ⁵ Conti, L. M.; Herdies, D. L.; Alvim, D. S.; Corrêa, S. M.; Aerosol Air Qual. Res. 2022, 22, 210364. [Crossref]
  » Crossref
• ⁶ Alvim, D. S.; Herdies, D. L.; Corrêa, S. M.; Basso, L. S.; Khalid, B.; Silva, G. F. P.; Oyerinde, G.; de Carvalho, N. A.; Coelho, S. M. S. C.; Figueroa, S. N.; Remote Sens. 2023, 15, 1262. [Crossref]
  » Crossref
• ⁷ Conselho Nacional do Meio Ambiente (CONAMA); Resolução CONAMA No. 18, de 6 de maio de 1986, Dispõe sobre a Criação do Programa de Controle de Poluição do Ar por Veículos Automotores - PROCONVE; Publicada no Diário Oficial da União (DOU), de 17 de junho de 1986, Seção 1, p. 8792-8795. [Link] accessed in October 2025
  » Link
• ⁸ Atkinson, R.; Atmos. Environ. 2000, 34, 2063. [Crossref]
  » Crossref
• ⁹ Seinfeld, J. H.; Pandis, S. N.; Atmospheric Chemistry and Physics: from Air Pollution to Climate Change, 3^rd ed.; Wiley: USA, 2016.
• ¹⁰ WHO Regional Office for Europe, Air Quality Guidelines - Global Update 2005. Particulate Matter, Ozone, Nitrogen Dioxide and Sulfur Dioxide; Copenhagen: WHO Regional Office for Europe, 2006. [Link] accessed in Ocotber 2025
  » Link
• ¹¹ World Health Organization (WHO); Global Air Quality Guidelines: Particulate Matter (PM2.5 and PM10), Ozone, Nitrogen Dioxide, Sulfur Dioxide and Carbon Monoxide; WHO: Geneva, 2021. [Link] accessed in October 2025
  » Link
• ¹² Agência Nacional do Petróleo, Gás Natural e Biocombustíveis (ANP), Painel Dinâmico de Produtores de Biodiesel, https://www.gov.br/anp/pt-br/centrais-de-conteudo/paineis-dinamicos-da-anp/paineis-e-mapa-dinamicos-de-produtores-de-combustiveis-e-derivados/painel-dinamico-de-produtores-de-biodiesel, accessed in October 2025.
  » https://www.gov.br/anp/pt-br/centrais-de-conteudo/paineis-dinamicos-da-anp/paineis-e-mapa-dinamicos-de-produtores-de-combustiveis-e-derivados/painel-dinamico-de-produtores-de-biodiesel
• ¹³ Ministério dos Transportes, Frota de Veíclulos 2023, https://www.gov.br/transportes/pt-br/assuntos/transito/conteudo-Senatran/frota-de-veiculos-2023, accessed in October 2025.
  » https://www.gov.br/transportes/pt-br/assuntos/transito/conteudo-Senatran/frota-de-veiculos-2023
• ¹⁴ Instituto Brasileiro de Petróleo e Gás (IBP), Vendas Anuais de Combustíveis, https://www.ibp.org.br/hub-de-conhecimento/observatorio-do-setor/snapshots/maiores-produtores-mundiais-de-petroleo-em-2023/, accessed in October 2025.
  » https://www.ibp.org.br/hub-de-conhecimento/observatorio-do-setor/snapshots/maiores-produtores-mundiais-de-petroleo-em-2023/
• ¹⁵ Nogueira, T.; Dominutti, P. A.; de Carvalho, L. R. F.; Fornaro, A.; Andrade, M. F.; Fuel 2014, 134, 505. [Crossref]
  » Crossref
• ¹⁶ Ghazali, W. N. M. W.; Mamat, R.; Masjuki, H. H.; Najafi, G.; Renewable Sustainable Energy Rev. 2015, 51, 585. [Crossref]
  » Crossref
• ¹⁷ Presidência da República, Secretaria-Geral; Lei No. 13.263 de 23 de março de 2016, altera a Lei No. 13.033, de 24 de setembro de 2014, para Dispor sobre os Percentuais de Adição de Biodiesel ao Óleo Diesel Comercializado no Território Nacional; Diário Oficial da União (DOU), Brasília, Brazil, 2016. [Link] accessed in October 2025
  » Link
• ¹⁸ Agência Nacional do Petróleo, Gás Natural e Biocombustíveis (ANP), https://www.gov.br/anp/pt-br/centrais-de-conteudo/dados-abertos/arquivos/pb-da-biodiesel.zip, accessed in October 2025.
  » https://www.gov.br/anp/pt-br/centrais-de-conteudo/dados-abertos/arquivos/pb-da-biodiesel.zip
• ¹⁹ Demirbas, A.; Energy Convers. Manage. 2009, 50, 14. [Crossref]
  » Crossref
• ²⁰ Ribas, W. F.; Bilotta, P.; Janissek, P. R.; Carvalho Filho, M. A. S.; Penteado Neto, R. A.; Eng. Sanit. Ambiental 2016, 21, 437. [Crossref]
  » Crossref
• ²¹ ABNT NBR 12026/2009: Light Road Vehicles - Determination of Aldehydes and Ketones in Exhaust Gas by Liquid Chromatography - DNPH Method; Brazil, 2009. [Link] accessed in October 2025
  » Link
• ²² ChemService, COA Search, https://us.store.analytichem.com/pages/coa-search, accessed in October 2025.
  » https://us.store.analytichem.com/pages/coa-search
• ²³ Dr. Ehrenstorfer, https://www.lgcstandards.com/BR/pt/Dr-Ehrenstorfer/cat/279845, accessed in October 2025.
  » https://www.lgcstandards.com/BR/pt/Dr-Ehrenstorfer/cat/279845
• ²⁴ Barnes I.; Rudzinski, K. J.; Environmental Simulation Chambers: Application to Atmospheric Chemical Processes; Springer: Poland, 2006.
• ²⁵ Agência Nacional do Petróleo, Gás Natural e Biocombustíveis (ANP); Resolução No. 807/2020, Estabelece a Especificação da Gasolina de Uso Automotivo e as Obrigações quanto ao Controle da Qualidade a Serem Atendidas pelos Agentes Econômicos que Comercializarem o Produto em Todo o Território Nacional; ANP: Brasília, 2020. [Link] accessed in October 2025
  » Link
• ²⁶ ABNT NBR 6601/2021: Light Duty Road Vehicles - Determination of Hydrocarbons, Carbon Monoxide, Nitrogen Oxide, Carbon Dioxide and Particulate Material on Exhaust Gas; Brazil, 2021. [Link] accessed in October 2025
  » Link
• ²⁷ Dayane Netto, F.; Chedid, J. P.; Daemme, L. C.; Penteado Neto, R. A.; Corrêa, S. M.; e Souza, T. C.; Dantas, T. C.; SAE Tec. Pap. 2021, 2020-36-0222. [Crossref]
  » Crossref
• ²⁸ Oliveira, A. F. G.; Revista Eletronica Nutritime 2008, 5, 777. [Link] accessed in October 2025
  » Link
• ²⁹ Larson, B.; Farber, R.; Applied Statistics, 6^th ed.; Pearson: São Paulo, 2015.
• ³⁰ Agarwal, A. K.; Shrivastava, A.; Prasad, R. K.; Renewable Energy 2016, 99, 564. [Crossref]
  » Crossref
• ³¹ Man, X. J.; Cheung, C. S.; Ning, Z.; Wei, L.; Huang, Z. H.; Fuel 2016, 180, 41. [Crossref]
  » Crossref
• ³² Amaral, B. S.; Ventura, L. M. B.; Amaral, A. S.; Neto, F. R. A.; Gioda, A.; J. Braz. Chem. Soc. 2017, 28, 659. [Crossref]
  » Crossref
• ³³ Turrio-Baldassarri, L.; Battistelli, C. L.; Conti, L.; Crebelli, R.; De Berardis, B.; Iamiceli, A. L.; Gambino, M.; Iannaccone, S.; Sci. Total Environ. 2004, 327, 147. [Crossref]
  » Crossref
• ³⁴ Guarieiro, L. L. N.; Vasconcellos, P. C.; Solci, M. C.; Rev. Virtual Quim. 2011, 3, 434. [Crossref]
  » Crossref
• ³⁵ Martins, E. M.; Arbilla, G.; Bauerfeldt, G. F.; de Paula, M.; Chemosphere 2007, 67, 2096. [Crossref]
  » Crossref
• ³⁶ Karavalakis, G.; Gysel, N.; Schmitz, D. A.; Cho, A. K.; Sioutas, C.; Schauer, J. J.; Cocker, D. R.; Durbin, T. D.; Sci. Total Environ. 2017, 584-585, 1230. [Crossref]
  » Crossref
• ³⁷ Fontaras, G.; Karavalakis, G.; Kousoulidou, M.; Tzamkiozis, T.; Ntziachristos, L.; Bakeas, E.; Stournas, S.; Samaras, Z.; Fuel 2009, 88, 1608. [Crossref]
  » Crossref
• ³⁸ George, I. J.; Hays, M. D.; Snow, R.; Faircloth, J.; George, B. J.; Long, T.; Baldauf, R. W.; Environ. Sci. Technol. 2014, 48, 14782. [Crossref]
  » Crossref
• ³⁹ Karavalakis, G.; Johnson, K. C.; Hajbabaei, M.; Durbin, T. D.; Fuel 2016, 181, 259. [Crossref]
  » Crossref
• ⁴⁰ Magara-Gomez, K. T.; Olson, M. R.; Okuda, T.; Walz, K. A.; Schauer, J. J.; Atmos. Environ. 2012, 50, 307. [Crossref]
  » Crossref
• ⁴¹ Ratcliff, M. A.; Dane, A. J.; Williams, A.; Ireland, J.; Luecke, J.; McCormick, R. L.; Voorhees, K. J.; Environ. Sci. Technol. 2010, 44, 8343. [Crossref]
  » Crossref
• ⁴² de Souza, C. V.; Corrêa, S. M.; Atmos. Environ. 2015, 103, 222. [Crossref]
  » Crossref
• ⁴³ de Souza, C. V.; Corrêa, S. M.; Fuel 2016, 185, 925. [Crossref]
  » Crossref
• ⁴⁴ Bamford, H. A.; Baker, J. E.; Atmos. Environ. 2003, 37, 2077. [Crossref]
  » Crossref
• ⁴⁵ Finlayson-Pitts, B. J.; Pitts, J. N.; Global Tropospheric Chemistry and Climate Change; Academic Press: San Diego, USA, 2000.
• ⁴⁶ Eiguren-Fernandez, A.; Miguel, A. H.; Lu, R.; Purvis, K.; Grant, B.; Mayo, P.; Di Stefano, E.; Cho, A. K.; Froines, J.; Atmos. Environ. 2008, 42, 2312. [Crossref]
  » Crossref
• ⁴⁷ Dobbins, R. A.; Fletcher, R. A.; Benner Jr., B. A.; Hoeft, S.; Combust. Flame 2006, 144, 773. [Crossref]
  » Crossref
• ⁴⁸ California Air Resources Board (CARB); SOP MLD144 Rev. 1.0: Procedure for the Determination of Polynuclear Aromatic Hydrocarbons in Particulate Matter using Gas Chromatography/Mass Spectrometry; CARB: USA, 2019. [Link] accessed in October 2025
  » Link

Edited by

• Editor handled this article:
  Andrea R. Chaves (Executive)

Publication Dates

• Publication in this collection
  28 Nov 2025
• Date of issue
  2025

History

• Received
  13 Aug 2025
• Published
  20 Oct 2025