Urban Tree Species Capturing Anthropogenic Volatile Organic Compounds—Impact on Air Quality

Abstract

Tropospheric ozone (O₃) and other pollutants significantly affect Chile’s Metropolitan Region, posing risks to human health. As a secondary pollutant and a major photochemical oxidant, O₃ formation is driven by anthropogenic volatile organic compounds (AVOCs) from the residential and transport sectors, the main sources of gaseous emissions. This study evaluated the AVOC capture capacity of leaf material from two tree species, Quillaja saponaria (native species) and Robinia pseudoacacia (exotic species), as potential urban biomonitors. Leaf samples were collected near nine SINCA official monitoring stations and the Antumapu University Campus, stored frozen, and analyzed by HS-SPME-GC/MSD for AVOC quantification. Photochemical reactivity and O₃ formation potential were assessed using equivalent propylene concentration (Prop-Equiv) and Ozone Formation Potential (OFP) methods. The results showed that both species captured atmospheric AVOCs, confirming their role as bioindicators. However, Q. saponaria adsorbed significantly higher AVOC concentrations and exhibited greater tropospheric O₃ formation potential than R. pseudoacacia. Given the AVOC adsorption capacity of both tree species, they could be used as biomonitors for styrene and also as a biomonitor for toluene in the case of Q. saponaria. This research highlights the importance of selecting tree capacity to improve urban air quality.

1. Introduction

The United Nations [1] reports that 56% of the global population currently resides in urban areas, with a projection to rise to 68% by 2050, underscoring the urgency of addressing air quality issues in densely populated areas.

In urban areas, vehicle emissions are the primary sources of anthropogenic volatile organic compounds (AVOCs) [2,3]. Gases containing lineal and aromatic compounds are released into the atmosphere, many of which are added to gasoline as antiknock agents, replacing lead [4,5]. Consequently, vehicles are a significant source of tropospheric ozone (O₃) precursors, including both volatile organic compounds (VOCs) and inorganic gases such as nitrogen oxides (i.e., NO + NO₂ = NOx).

Santiago is the political and administrative capital of Chile (33.5° S, 70.6° W, 500–800 m ASL) [6] and one of the five provinces of the Metropolitan Region (MR). The southeastern Pacific subtropical anticyclone, located around 30° S and 90° W, worsens conditions by creating an extensive area of atmospheric subsidence. The city’s unique geography, surrounded by mountains (between 2000 and 5000 m ASL), traps pollutants and limits their dispersion, further aggravating air quality issues [7,8,9]. Air quality in the MR is especially compromised by high levels of photochemical O₃ during the austral spring and summer seasons and by particulate matter (PM) in autumn and winter [10,11]. Photochemical pollutants such as NOx and O₃, along with VOCs and PM, contribute to the city’s poor air quality. Chilean regulations for O₃ are exceeded on several days during the spring–summer season in the eastern part of the MR [10,12], thereby increasing the population’s exposure to photochemical smog [13,14,15,16,17].

AVOCs, primarily produced by human activity, represent the third largest contribution to gaseous emissions in the MR, with emissions largely dominated by transport and residential sectors. Transportation accounts for 61.1% of Santiago’s pollution, while residential firewood use contributes 37.1% [18]. Additionally, VOC emissions reach 7754.5 t/year [19], and vehicles emit approximately 32,924 tons of NOx annually [20].

The exhaust from vehicles contains various pollutants, notably PM (soot of various sizes), carbon monoxide (CO), NOx, VOCs, and heavy metals such as lead, cadmium, arsenic, and chromium with concentrations reported in the range of 0.60–205.00 ng/m³ in urban environments [21]. The AVOCs of greatest interest are benzene and its derivatives and polycyclic aromatic species (PAHs); some of these have toxic, teratogenic, mutagenic, and/or carcinogenic effects, causing serious damage to human health and plant growth [22,23]. These emissions are caused by the gasification or evaporation of petroleum-based or other organic substances through the burning of fossil fuels and evaporation from paints, glues, and solvents, among other important sources [2,3,24,25]. Gasoline, a major contributor to these emissions in urban areas, contains a complex mixture of approximately 1500 compounds, including paraffin (alkanes), olefins (alkenes), cycloalkanes, and aromatic hydrocarbons such as benzene, toluene, ethylbenzene, and xylene (BTEX) [26,27]. Benzene causes immune cells to produce excessive inflammatory factors, leading to inflammatory responses and damage to the bone marrow and other hematopoietic tissues. Additionally, benzene and its metabolites activate the intrinsic apoptotic pathway within cells, promoting the programmed cell death of hematopoietic cells and causing hematotoxicity [28].

The multiple health effects of AVOCs have led researchers to investigate different strategies to rapidly remove emissions or reduce their concentrations in the air using thermal oxidation and photocatalytic oxidation [29,30,31,32].

In this study, we propose a natural approach using plants, which offer an eco-friendly and cost-effective alternative to conventional physical and chemical air purification methods [33]. Plants contribute to environmental remediation by absorbing and accumulating air pollutants through their leaf surfaces. Additionally, they enhance ecological balance by cycling CO₂ and O₂. Certain plant species react distinctly to specific air pollutants, serving as useful indicators of air quality. Various authors [12,34,35,36,37] have reported on the atmospheric purification capacities of plant species, particularly in urbanized environments. Their research highlights key functions, such as CO₂ capture through photosynthesis, the ability of leaves to retain PM, and the mitigation of urban heat island effects, among other benefits known as ecosystem services, which also includes improving both mental and physical health [38,39,40,41]. In addition, trees have been shown to play an important role in the removal of nitrogen compounds from the atmosphere [42]. However, there are few local studies on the uptake of gaseous air pollutants by the leaves of trees exposed to defined pollution sources, and specific information for AVOCs is limited [43].

Photochemical ozone formation is a nonlinear process influenced by the relative concentrations of VOCs and NOx, which divides O₃ formation into two categories: VOC-limited and NOx-limited. Determining whether O₃ formation is limited is essential for developing effective control strategies. In VOC-limited conditions, where O₃ formation is sensitive to changes in VOC levels, it is possible to estimate each VOC’s contribution to O₃ production. This approach assumes that all VOCs in the atmosphere react with hydroxyl (OH) radicals, thereby contributing to O₃ formation [44]. Further data show that primary emissions of NOx and VOCs from motor vehicle exhausts continue to be the main driver of photochemical air pollution in the MR [45]. However, urban vegetation also emits biogenic volatile organic compounds (BVOCs), which are highly reactive and contribute to O₃ formation due to their ecophysiological functions [2,10,46]. BVOCs include alkanes, alkenes, carbonyls, alcohols, esters, ethers, and acids [47], of which terpenes are the most important. The basic molecule of terpenes is isoprene (2-methyl-1,3-butadiene, C₅H₈), a highly reactive compound due to its double bonds.

On the other hand, high concentrations of tropospheric O₃ have significant adverse effects on vegetation as they penetrate plant stomata and trigger the formation of reactive oxygen species (ROS), leading to membrane damage, reduced photosynthetic capacity, and lowering crop yields [48]. The VOC emissions (precursors of O₃) are the third source of gaseous emissions to the atmosphere of the MR, after CO₂ and CO. In recent years, the MR has experienced a significant increase in its vehicle fleet; in 2022, it had 2,235,482 vehicles, of which 77.2% were petrol vehicles and 22.3% were diesel vehicles [49]. There is no specific information on benzene, toluene, and xylenes (BTX) or other benzene derivatives [18].

Globally, research on the potential for the adsorption and/or absorption of gaseous air pollutants by the leaves of trees exposed to pollution sources has been growing [50,51,52]. Some studies have demonstrated the ability of plants to mitigate various VOCs [53,54]. While research on AVOCs specifically is less extensive compared to research on VOCs in general, emerging evidence highlights the importance of plants in mitigating their impacts, especially in urban and indoor environments [37,52,53,54,55]. To quantify the impact of AVOCs retained on leaf surfaces on tropospheric O₃ formation, estimators of photochemical reactivity, such as propylene equivalents (Prop-equiv) and Ozone Formation Potential (OFP), are commonly used. These parameters help assess the contribution of AVOCs to atmospheric chemistry by measuring their reaction with OH radicals and other reactive species [43].

Trees remove or retain AVOCs through processes determined by both the anatomical structure of their leaves and complex physiological mechanisms. The mechanisms for AVOC removal and retention by leaves include absorption through stomata and subsequent metabolic degradation, as well as non-stomatal adsorption, which may depend on the composition of the leaf cuticle [55]. This study focuses exclusively on the adsorption mechanism.

Establishing the capture of O₃ precursors AVOCs, as a quantifiable parameter, expressed in terms of Prop-equiv and the OFP from compounds retained in leaf material, allows for the assessment of their O₃ formation capacities. This provides a representative estimation of the potential of different tree species to mitigate O₃ formation at the tropospheric level. It will also help to identify which tree species are most effective for introduction into urban afforestation projects or for use as bioindicators of photochemical pollution.

This work estimates the capacity of the leaf material from two tree species, Quillaja saponaria (native species) and Robinia pseudoacacia (exotic species), to capture AVOCs in urban environments.

2. Materials and Methods

2.1. Tree Species Selection

The Metropolitan Region is located in an area of high plant endemism; however, the urban forest of the MR is composed mainly of exotic tree species: more than 90% are exotic to the country, representing 84–96% of the total number of trees [56]. The species selected in this work were Q. saponaria (3.48% abundance) and R. pseudoacacia (6.4% abundance) in 2014.

Quillaja saponaria (Spanish name Quillay) [57,58] is an evergreen tree that grows from 29.9533° to 38.5400° latitude S. It is a medium-growing species (up to 15 m in height) with a longevity of approximately 100 to 150 years. Q. saponaria has interesting physiological characteristics, such as tolerating different environmental conditions for its growth and being able to develop in nutrient-poor soils, in areas with high thermal oscillations, and in extreme drought conditions [59,60]. These attributes are adaptations of the species to the seasonal contrasts of the Mediterranean climate, such as intense temperatures in summer, but also frost, sunstroke, and low water availability [61,62]. The leaves of Q. saponaria are simple, alternate, smooth, and ovate.

Robinia pseudoacacia is a species of the Fabaceae family. It is native to the eastern United States of America, although it has been introduced to other countries. It is characterized by its great capacity to recover soils and control erosion, its rapid growth (reaching 25 m in height), and its tolerance to pollution. It is a feral species in the Chilean countryside, especially in the central area of the country. It is characterized by a trunk with gray, cracked bark that grows twisted, forked, and full of branches with a wide cup with an ovate and asymmetrical shape. The leaves, with petiole and rachis with 5 acute ribs, are 10–25 cm long with 9–19 leaflets of 2–5 by 1.5–3 cm [63].

2.2. Sampling Sites

Sampling was conducted at locations near air quality monitoring stations of the National Air Quality Information System (SINCA), including communes Cerrillos, Cerro Navia, El Bosque, Independencia, La Florida, Las Condes, Parque O’Higgins, Puente Alto, Quilicura, and the Antumapu Campus of the University of Chile at La Pintana. For each site, composite leaf samples were collected from 10 individuals of R. pseudoacacia and 10 individuals of Q. saponaria [64]. Approximately 30 g of biomass per species was collected using a perimeter sampling approach due to the maximum height of the individuals (1.6 m). The sampling campaign was carried out during the austral summer of 2017 (January) and funded by the Fondecyt N°1140319 grant, Conicyt, Chile. Samples were immediately preserved on dry ice and stored at −20 °C until chemical analysis. The sampling locations are indicated in Figure 1.

2.3. Analytical Treatment

Leaf samples were milled using a cryogenic mill (Retsch, Haan, Germany). Details in Supplementary Material Table S1. Each 2 g sample was placed in a 20 mL glass vial for Headspace with PTFE/silicone septum. Headspace Solid-Phase Microextraction (HS-SPME) extraction [65,66] was performed using a GC CombiPAL 80 Agilent containing a fiber assembly with 65 μm Polydimethylsiloxane/Divinylbenzene (PDMS/DVB) coating (Supelco, Bellefonte, PA, USA). Details in Supplementary Material Table S2. The fiber was conditioned according to the manufacturer’s instructions in the GC injection port at 250 °C for 30 min before use. The analytes were extracted for 50 min at 70 °C, and desorption occurred for 5 min at 250 °C in the injection port. A blank fiber experiment was conducted to ensure the absence of contaminants in the fiber. Intermediate standard solutions were prepared from a volatile compound mixture #2 [67] to construct the calibration curve, covering a working range of 0.05 ng/g to 50 ng/g. Each sample (including blanks, duplicates, quality controls, and calibration points) received a 100 µL aliquot of a 10 µg/L ααα-trifluorotoluene solution internal standard [67]. This internal standard was used to correct for variability in sample preparation and instrumental analysis.

Sample analysis was conducted with an Agilent Technologies 7890A GC (Agilent Technologies, Inc., Santa Clara, CA, USA) coupled to an Agilent Technologies 5975C mass spectrometer (MS) in Selective Ion Monitoring (SIM) mode. Splitless injection was performed onto an HP-5MS capillary column (30 m × 0.25 mm i.d., 0.25 μm film thickness) with helium (99.999%) as the carrier gas at a flow rate of 1 mL/min. The injection port temperature was set to 250 °C, and the column temperature was programmed to start at 35 °C. The mass selective detector (MS) operated at an ionization voltage of 70 eV. The programming conditions for the equipment are provided in Supplementary Material Table S3, which assisted in reducing noise effects and increasing signal gain.

Table 1. Transitions corresponding to the quantifier and qualifier ions (m/z) along with the retention times for each analyzed compound.

Analyte	CAS No.	Qualifier Ion Q (m/z)	Qualifier Ion q1 (m/z)	Qualifier Ion q2 (m/z)	Retention Time (min)
cis-1,2-Dichloroethylene	156-59-2	61	96	98	2.13
trans-1,2-Dichloroethylene	156-60-5	61	96	98	2.43
cis-1,3-Dichloropropene	1061-01-5	75	77	110	4.72
trans-1,3-Dichloropropene	1061-02-6	75	77	110	4.72
Benzene	71-43-2	78	77	51	3.06
Bromobenzene	108-86-1	77	156	158	12.34
Bromochloromethane	74-97-5	49	130	128	2.54
Bromodichloromethane	75-27-4	83	85	47	3.84
sec-Butylbenzene	135-98-8	105	120	77	15.45
tert-Butylbenzene	98-06-6	119	91	41	15.11
Chlorobenzene	108-90-7	112	77	114	8.47
2-Chlorotoluene	95-49-8	91	126	89	13.17
4-Chlorotoluene	106-43-4	126	128	89	13.43
1,2-Dibromo-3-chloropropane	96-12-8	157	75	155	16.38
Dibromochloromethane	124-48-1	127	129	131	6.53
1,2-Dibromoethane	106-93-4	107	109	93	6.91
Dibromomethane	74-95-3	93	174	95	3.73
1,1-Dichloro-1-propene	563-58-6	75	39	110	2.98
1,2-Dichlorobenzene	95-50-1	146	148	111	15.86
1,3-Dichlorobenzene	541-73-1	146	148	111	15.39
1,4-Dichlorobenzene	106-46-7	146	148	111	15.53
1,1-Dichloroethane	75-34-3	63	65	83	2.22
1,2-Dichloroethane	107-06-2	62	64	49	2.90
1,1-Dichloroethylene	75-35-4	61	96	98	1.90
Dichloromethane	75-09-2	49	84	86	1.97
1,2-Dichloropropane	78-87-5	63	62	41	3.68
1,3-Dichloropropane	142-28-9	76	41	78	6.22
2,2-Dichloropropane	594-20-7	77	41	79	2.52
Ethylbenzene	100-41-4	91	106	51	9.18
Hexachloro-1,3-butadiene	87-68-3	225	227	223	17.56
Isopropylbenzene	98-82-8	105	120	77	12.22
4-Isopropyltoluene	99-87-6	119	134	91	15.70
Naphthalene	91-20-3	128	102	127	17.31
n-Propylbenzene	103-65-1	120	65	78	13.35
Styrene	100-42-5	104	78	51	10.69
1,1,1,2-Tetrachloroethane	630-20-6	131	133	117	8.65
1,1,2,2-Tetrachloroethane	79-34-5	83	85	61	11.85
Tetrachloroethylene	127-18-4	166	164	131	7.05
Tetrachloromethane	56-23-5	117	119	121	3.06
Toluene	108-88-3	91	92	65	5.55
Tribromomethane	75-25-2	173	171	175	10.24
1,2,3-Trichlorobenzene	87-61-6	180	182	74	17.25
1,2,4-Trichlorobenzene	120-82-1	180	145	184	17.24
1,1,1-Trichloroethane	71-55-6	97	99	61	2.85
1,1,2-Trichloroethane	79-00-5	97	83	99	5.75
Trichloroethylene	79-01-6	95	130	132	3.69
Trichloromethane	67-66-3	83	85	47	2.54
1,2,3-Trichloropropane	96-18-4	75	105	77	12.06
α,α,α-Trifluorotoluene (EI)	98-08-8	146	127	96	3.96
1,2,4-Trimethylbenzene	95-63-6	105	120	77	13.98
1,3,5-Trimethylbenzene	108-67-8	105	120	77	13.50
m-Xylene	108-38-3	91	106	105	9.58
o-Xylene	95-47-6	91	106	105	10.80
p-Xylene	106-42-3	91	106	105	9.58

summarizes the quantifier and qualifier ions for each analyzed compound.

2.4. Statistical Analysis

Comparative strategies were established in relation to a direct source of pollution or among tree species exposed to similar conditions of interaction with the pollution source. These strategies correspond to relative frequency distributions of each value in a dataset, expressed as a percentage of total frequencies.

Statistical evaluation was performed on the data to determine whether statistically significant differences existed among the studied tree species. This analysis was supported by p-value tests at a 95% confidence level, utilizing the Student’s t-test to assess a significant difference between the mean values of the two groups.

2.5. Estimation of Photochemical Reactivity

In urban environments and particularly in this work, the sources of VOC emissions are physically close, and the formation of photochemical O₃ is limited and sensitive to any changes in VOC concentrations [44,45,68]. Therefore, equations can be applied to quantify the role of each VOC on O₃ formation, considering that all VOCs undergo atmospheric reactions with OH radicals, leading to O₃ production. The estimators of the photochemical reactivity of VOCs and/or O₃ formation potential include calculations of Propylene Equivalent Concentration (Prop-Equiv) [69] and Ozone Formation Potential (PFO) [24,70,71].

To calculate the Propylene Equivalent Concentration for each individual VOC, the following equation was used:

Prop-Equiv_(i) = conc_(i) × K_OH(i)/K_OH(C3H6)

(1)

where Prop-Equiv_(i) is the Propylene Equivalent Concentration; conc_(i) is the concentration of AVOC_i; K_OH(i) is the reaction rate constant of AVOC_i with the OH radical; and K_OH(C3H6) is the reaction rate constant of propylene with the OH radical [72].

The Ozone Formation Potential (OFP) for individual AVOCs, using the maximum incremental reactivity (MIR) method, is described in Equation (2):

OFP_(i) = conc_(i) × MIRcoef_(i)

(2)

where OFP_(i) is the Ozone Formation Potential of the AVOC species, in µg/m³; conc_(i) is the AVOC concentration expressed in ppbC; and MIRcoef_(i) is the maximum incremental reactivity coefficient of compound i, expressed in gO₃/gVOC.

Consequently, photochemical O₃ formation should decrease with the reduction in VOC concentrations. Thus, it can be inferred that VOCs retained in foliar material are no longer available to interact in the environment, thereby reducing tropospheric O₃ formation. The elimination of O₃-generating precursors (AVOC) was established as a quantifiable parameter, expressed in terms of Prop-equiv and the OFP. Based on the compounds retained in foliar material, their O₃ formation capacities are determined, thus providing a representative estimate of the tree species’ potential to mitigate tropospheric O₃ formation [43].

2.6. Reactivity Control Index

A balanced Reactivity Control Index (RCI) was defined by normalizing the previously described expressions to identify the chemical species with the highest reactivity impact, as shown in Equation (3):

$${\mathrm{R}\mathrm{C}\mathrm{I}}_{\left(\mathrm{i}\right)}=K_1\times {\displaystyle \frac{\mathit{\text{Prop-equiv}}\left(i\right)-\mathit{\text{Prop-equiv}}\left(min\right)}{\mathit{\text{Prop-equiv}}\left(max\right)-\mathit{\text{Prop-equiv}}\left(min\right)}}+K_2\times {\displaystyle \frac{OFP\left(i\right)-OFP\left(min\right)}{OFP\left(max\right)-OFP\left(min\right)}}$$

(3)

where RCI_(i) is the normalized Reactivity Control Index for AVOC species I and k₁ and k₂ are weighting factors, each set to 0.5. Prop-equiv(min) and Prop-equiv(max) are the minimum and maximum Prop-equiv_(i) values among the AVOC species. OFP_(min) and OFP_(max) are the minimum and maximum OFP(i) values among the AVOC species studied [73].

3. Results

3.1. Quantification of the Anthropogenic Volatile Organic Compounds in the Foliar Material of the Two Tree Species

The following VOCs were quantified: ethylbenzene, styrene, toluene, 1,3,5-trimethylbenzene (1,3,5-TMB), 1,3-dichlorobenzene, 4-isopropyltoluene, o-xylene, m/p-xylene, and n-propylbenzene (

Table 2. Concentrations of volatile organic compounds in leaf samples of R. pseudoacacia and Q. saponaria collected near each environmental monitoring station of the National Air Quality Information System (SINCA).

Sampling Site	Coordinatess		1,3-Dichlorobenzene		Ethylbenzene		4-Isopropyltoluene
			Q. saponaria	R. pseudoacacia	Q. saponaria	R. pseudoacacia	Q. saponaria	R. pseudoacacia
	Latitude	Longitude	Value (ng of C)		Value (ng of C)		Value (ng of C)
Antumapu	−33.570876	−70.633697	<LOQ	0.32 ± 0.03	<LOQ	<LOQ	<LOQ	0.52 ± 0.02
Cerrillos	−33.492879	−70.719397	<LOQ	0.31 ± 0.05	4.08 ± 0.03	3.55 ± 0.19	<LOQ	0.66 ± 0.02
Cerro Navia	−33.433088	−70.732082	<LOQ	0.36 ± 0.03	2.70 ± 0.03	2.70 ± 0.13	<LOQ	0.61 ± 0.03
El Bosque	−33.546913	−70.666728	<LOQ	0.33 ± 0.03	2.43 ± 0.03	4.25 ± 0.13	<LOQ	0.64 ± 0.03
Independencia	−33.422242	−70.651155	<LOQ	0.40 ± 0.04	<LOQ	0.50 ± 0.05	0.32 ± 0.05	0.73 ± 0.01
La Florida	−33.516623	−70.588089	<LOQ	0.13 ± 0.03	<LOQ	0.90 ± 0.05	<LOQ	0.30 ± 0.03
Las Condes	−33.377673	−70.52326	<LOQ	0.22 ± 0.03	<LOQ	0.05 ± 0.05	<LOQ	0.70 ± 0.04
Parque O’Higgins	−33.463379	−70.658585	<LOQ	0.46 ± 0.03	1.26 ± 0.06	3.61 ± 0.60	<LOQ	0.57 ± 0.04
Puente Alto	−33.591361	−70.594768	<LOQ	0.51 ± 0.03	1.95 ± 0.03	5.91 ± 0.15	<LOQ	0.70 ± 0.03
Quilicura	−33.34966	−70.723693	<LOQ	0.20 ± 0.09	< LOQ	2.71 ± 0.15	<LOQ	0.70 ± 0.03
Sampling site	Coordinates		n-Propylbenzene		Styrene		1,3,5-TMB
			Q. saponaria	R. pseudoacacia	Q. saponaria	R. pseudoacacia	Q. saponaria	R. pseudoacacia
	Latitude	Longitude	Value (ng of C)		Value (ng of C)		Value (ng of C)
Antumapu	−33.570876	−70.633697	<LOQ	0.46 ± 0.01	58.19 ± 1.43	45.08 ± 1.43	<LOQ	<LOQ
Cerrillos	−33.492879	−70.719397	0.28 ± 0.05	0.62 ± 0.01	50.30 ± 8.46	67.57 ± 8.46	0.21 ± 0.05	0.27 ± 0.025
Cerro Navia	−33.433088	−70.732082	<LOQ	0.61 ± 0.02	60.68 ± 0.74	57.54 ± 2.54	0.23 ± 0.05	0.20 ± 0.043
El Bosque	−33.546913	−70.666728	<LOQ	0.37 ± 0.13	52.81 ± 1.03	37.37 ± 0.34	<LOQ	0.20 ± 0.043
Independencia	−33.422242	−70.651155	<LOQ	0.70 ± 0.01	39.51 ± 1.75	70.35 ± 1.78	<LOQ	0.33 ± 0.034
La Florida	−33.516623	−70.588089	0.50 ± 0.10	0.13 ± 0.05	77.18 ± 8.21	25.54 ± 0.54	<LOQ	<LOQ
Las Condes	−33.377673	−70.52326	0.31 ± 0.05	0.70 ± 0.03	60.70 ± 8.38	31.04 ± 0.93	0.19 ± 0.06	<LOQ
Parque O’Higgins	−33.463379	−70.658585	0.43 ± 0.02	0.70 ± 0.03	49.06 ± 3.21	64.11 ± 5.53	<LOQ	0.29 ± 0.049
Puente Alto	−33.591361	−70.594768	0.37 ± 0.01	0.73 ± 0.01	52.14 ± 1.64	101.04 ± 5.17	0.36 ± 0.02	0.36 ± 0.02
Quilicura	−33.34966	−70.723693	<LOQ	0.39 ± 0.14	72.91 ± 2.20	45.41 ± 2.12	0.14 ± 0.016	0.14 ± 0.016
Sampling site	Coordinates		Toluene		m/p-Xylene		o-Xylene
			Q. saponaria	R. pseudoacacia	Q. saponaria	R. pseudoacacia	Q. saponaria	R. pseudoacacia
	Latitude	Longitude	Value (ng of C)		Value (ng of C)		Value (ng of C)
Antumapu	−33.570876	−70.633697	34.11 ± 1.94	1.93 ± 0.37	4.29 ± 0.27	4.03 ± 0.27	<LOQ	0.99 ± 0.09
Cerrillos	−33.492879	−70.719397	21.97 ± 0.57	3.33 ± 0.02	4.27 ± 1.63	7.14 ± 0.20	2.19 ± 0.15	2.49 ± 0.15
Cerro Navia	−33.433088	−70.732082	38.12 ± 0.40	3.36 ± 0.26	5.50 ± 0.18	6.09 ± 0.28	<LOQ	2.01 ± 0.18
El Bosque	−33.546913	−70.666728	24.28 ± 1.02	1.99 ± 0.26	4.17 ± 0.17	5.27 ± 0.48	1.13 ± 0.09	1.48 ± 0.08
Independencia	−33.422242	−70.651155	18.64 ± 1.13	3.59 ± 0.25	3.48 ± 0.27	7.83 ± 0.15	2.89 ± 0.10	2.89 ± 0.10
La Florida	−33.516623	−70.588089	29.52 ± 6.01	6.01 ± 0.04	7.19 ± 1.64	1.38 ± 0.05	<LOQ	<LOQ
Las Condes	−33.377673	−70.52326	21.01 ± 2.28	2.94 ± 0.74	6.16 ± 1.60	2.21 ± 0.05	1.82 ± 0.03	2.88 ± 0.12
Parque O’Higgins	−33.463379	−70.658585	21.01 ± 2.28	2.28 ± 0.04	4.66 ± 0.71	7.31 ± 0.43	2.88 ± 0.12	2.88 ± 0.12
Puente Alto	−33.591361	−70.594768	10.21 ± 1.64	5.17 ± 0.21	5.94 ± 0.16	10.44 ± 0.37	1.64 ± 0.05	2.82 ± 0.15
Quilicura	−33.34966	−70.723693	34.20 ± 1.15	2.17 ± 0.39	6.61 ± 0.38	4.37 ± 0.36	1.50 ± 0.13	1.50 ± 0.13

Values expressed in ng of C; LOQ, limit of quantification (0.10 ng of C).

). Although other VOCs of toxicological interest, such as benzene, were detected, their concentrations were below the limit of quantification (LOQ). In both species, styrene showed the highest concentrations, with a maximum value of 77.18 ng of C in Q. saponaria (La Florida) and 101.04 ng of C in R. pseudoacacia (Puente Alto). Toluene was the second most predominant compound, with 34.20 ng of C in Quilicura (Q. saponaria) and 6.01 ng of C at La Florida (R. pseudoacacia). Ethylbenzene showed significant concentrations in R. pseudoacacia at Cerrillos with 3.55 ng of C and Cerro Navia with 2.70 ng of C. In Q. saponaria, that compound was only present at four sampling sites: 2.43 ng of C at El Bosque, 4.08 ng of C at Cerrillos, and lower concentrations at Parque O’Higgins and Puente Alto. Also, m/p-xylenes reached their highest concentrations at Puente Alto for both species, with 10.44 ng of C in R. pseudoacacia and 6.61 ng of C in Q. saponaria. Additionally, o-xylene generally exhibited relatively low values, with a maximum of 2.88 ng of C at Parque O’Higgins (R. pseudoacacia).

Regarding 1,3-dichlorobenzene and 4-isopropyltoluene, their concentrations were mostly below the LOQ, except for small amounts detected in R. pseudoacacia at points such as Independencia (0.73 ng of C for 4-isopropyltoluene) and Cerrillos (0.66 ng of C for 1,3-dichlorobenzene).

The percentage distribution of AVOCs retained by Q. saponaria and R. pseudoacacia (Figure 2) varied significantly between species and sampling sites. In Q. saponaria (Figure 2a), styrene and toluene dominated the concentrations, with higher proportions at sites such as La Florida and Quilicura, where styrene represented more than 70% of the total. Conversely, in R. pseudoacacia (Figure 2b), a greater diversity of retained AVOCs was observed, with significant contributions from ethylbenzene and m/p-xylenes at sites such as Puente Alto and Cerrillos. Notably, styrene was the predominant compound in both species, especially in R. pseudoacacia at Puente Alto, where it accounted for over 80% of the total. These results reflect differences in the adsorption capacity of the species, as well as potential variations in emission sources and AVOC exposure across the evaluated sites.

3.2. Variability in the Concentration of Anthropogenic Volatile Organic Compounds in the Foliar Material of the Two Tree Species

Statistically significant differences were observed between the total concentrations of AVOC content in the leaves of Q. saponaria and R. pseudoacacia. The mean total AVOC concentration in Q. saponaria was 95.2 ± 19.7 ng of C, while in R. pseudoacacia, it was 69.0 ± 29.7 ng of C (Figure 3). The statistical analysis confirmed this difference with Tobs = 2.32 and a p-value = 0.032, which is below the significance level of α = 0.05 (Tobs > Tcri; p-value < 0.05).

The AVOC concentrations in the leaves of Q. saponaria and R. pseudoacacia are shown for individual compounds across all sampling sites (Figure 4). For most compounds, R. pseudoacacia tends to exhibit slightly higher concentrations compared to Q. saponaria, except for toluene, where Q. saponaria shows significantly higher adsorption. The average toluene concentration was 30.3 ± 8.48 ng of C for Q. saponaria and 2.66 ± 1.27 ng of C for R. pseudoacacia. Similarly, styrene concentrations were comparable between the two species, with 57.3 ± 11.3 ng of C for Q. saponaria and 54.5 ± 22.5 ng of C for R. pseudoacacia, although the latter displayed greater variability.

For compounds with intermediate concentrations, such as ethylbenzene and m/p-xylenes, R. pseudoacacia showed higher adsorption. The average ethylbenzene concentration was 1.68 ± 0.369 ng of C for Q. saponaria and 3.14 ± 1.69 ng of C for R. pseudoacacia. Similarly, m/p-xylenes concentrations were 5.23 ± 1.22 ng of C for Q. saponaria and 5.61 ± 2.73 ng of C for R. pseudoacacia.

For low-concentration compounds, such as 1,3,5-trimethylbenzene, 1,3-dichlorobenzene, 4-isopropyltoluene, and n-propylbenzene, both species exhibited similar values with no significant differences. For example, the concentrations of n-propylbenzene were 0.378 ± 0.104 ng of C for Q. saponaria and 0.498 ± 0.202 ng of C for R. pseudoacacia. Additionally, 1,3,5-trimethylbenzene concentrations were nearly identical, averaging 0.203 ± 0.140 ng of C for Q. saponaria and 0.199 ± 0.0915 ng of C for R. pseudoacacia.

3.3. Ozone Formation Potentials and Reactivity Control Index

The estimated values for propylene equivalents (Prop-equiv) and Ozone Formation Potentials (OFPs) revealed that Q. saponaria exhibited higher total AVOC values (Prop-equiv = 141.1 ppbC, OFP = 286.3 ppbC) compared to R. pseudoacacia (Prop-equiv = 129.6 ppbC, OFP = 181.6 ppbC) (

Table 3. Mean Propylene-Equivalent Concentrations (Prop-equiv) and Ozone Formation Potentials (OFPs) (ppb C) of anthropogenic volatile organic compounds (AVOCs) for Quillaja saponaria and Robinia pseudoacacia sampled near the monitoring stations of the National Air Quality Information System (SINCA).

	Prop-Equiv(i), ppbC		OFP(i), ppbC
	Q. saponaria	R. pseudoacacia	Q. saponaria	R. pseudoacacia
1,3-Dichlorobenzene	0.0	0.0	0.0	0.1
Ethylbenzene	0.5	0.9	5.2	9.7
4-Isopropyltoluene	0.0	0.0	1.4	2.6
n-Propylbenzene	0.0	0.0	0.0	0.0
Styrene	128.8	122.4	101.2	96.2
1,3,5-TMB	0.5	0.5	2.8	2.7
Toluene	6.6	0.6	123.7	10.9
Xylenes	4.6	5.1	51.1	58.5
Total AVOC	141.1	129.6	286.3	181.6

). Styrene was the most important contributor to the reactivity control indices (RCIs) in both species, with the highest values observed in R. pseudoacacia (RCI = 1.00) and slightly lower in Q. saponaria (RCI = 0.90). Toluene showed marked differences between species, with significantly higher contributions in Q. saponaria (RCI = 6.6) compared to R. pseudoacacia (RCI = 0.6). Xylenes were more relevant for R. pseudoacacia (RCI = 5.1) than for Q. saponaria (RCI = 4.6). These trends are visually summarized in Figure 5, which highlights the compound-specific contributions to the RCI for each species. Other AVOCs, such as ethylbenzene, 1,3,5-trimethylbenzene (1,3,5-TMB), and 4-isopropyltoluene, displayed minimal contributions, while 1,3-dichlorobenzene and n-propylbenzene showed negligible reactivity in both species.

4. Discussion

Chile’s steady economic growth, concentrated in Santiago city (almost 7 million population), has resulted in a significant expansion in the car fleet, with over 6.2 million vehicles nationwide as of 2022 and around 36.3% concentrated in the Metropolitan Region. This growth, combined with increased fuel consumption, has contributed to severe air pollution challenges in the Region [49]. Rising private vehicle use and emissions of NOx and VOC are the most important factors producing photochemical pollution, with evidence showing a VOC-limited ozone formation regime [44,45]. Harmful ozone days remained high in the east and northeast areas of the MR [10,45].

Some plant species are highly sensitive to particular air pollutants and show specific responses to pollutant effects by showing specific damage symptoms. These species can be used to detect and monitor the presence or absence of air pollutants [74]. A study in Gothenburg, Sweden, measured PAHs in oak leaves and pine needles in 2018. The results showed that oak leaves had decreased low-molecular-mass PAHs (L-PAHs) but increased high-molecular-mass PAHs (H-PAHs) over time. Pine needles, especially older ones, accumulated more L-PAHs than oak leaves, but H-PAH concentrations were higher in oak leaves on a mass basis [75]. Robinia pseudoacacia was used to monitor trace elements in air and soil pollution [76]. Olea europaea was used as a monitor of trace elements in the air in cities of Turkey [77,78], and Betula pendula was used to study micro- and macro-elements in the air of Plovdiv, Bulgaria [79].

This study provides a detailed assessment of the presence of AVOCs in the range of gasoline and in the leaves of urban trees in the Metropolitan Region of Chile and highlights the importance of trees as indicators of urban air quality. The concentration profiles of AVOC detected in the leaf material of Q. saponaria and R. pseudoacacia evaluate, for the first time, to our knowledge, the ecological impact of AVOC emissions in urban environments and their potential contributions to tropospheric O₃ formation. The tree species R. pseudoacacia and Q. saponaria are located near each monitoring station of the SINCA network and are exposed to the same environmental conditions, reflecting the air quality of Santiago. Therefore, the results obtained in this study for each site retain these characteristics. The differences in concentration levels found between the two species may reflect their distinct species-specific responses to environmental stressors due to physiological traits, anatomical structures, or other factors that influence the adsorption of the kind of AVOC. The leaves of Q. saponaria are simple, alternate, smooth, and ovate. In contrast, R. pseudoacacia has composite leaves, with 10–25 cm long 9–19 leaflets of 2–5 by 1.5–3 cm.

Additionally, sampling was conducted at the Antumapu Campus of the University of Chile to increase the representativeness of the information for the Metropolitan Region (Figure 1) but showing important differences with the other sampled sites in relation to the number and concentration of quantified AVOC. An intra-species evaluation revealed homogeneity in the relative percentages. A particular situation was observed in the leaves of R. pseudoacacia from the Puente Alto station, which exhibited a high concentration of styrene (Table 3). We have no explanation at this time about this difference in Puente Alto, but the proportionality in its concentration profile was maintained, as higher values were also quantified for other chemical species.

Effects of atmospheric pollution by trace elements on R. pseudoacacia L. leaves were studied in the city of Isfahan, Iran [76]. Zhang et al. (2012) [80] demonstrate that relatively small trees (<15 m) have leaf morphologies that vary with height and that such variation depends on site moisture variability. Leaf area and stomata area decreased with height, while leaf mass per area, carbon isotope composition (δ ¹³C), and stomata density increased with height.

All the arboreal individuals studied in this work are clones (Fondecyt grant N°1140319) of similar height to avoid a factor of behavioral difference by morphology in terms of the adsorption of the quantified AVOC. However, Egas et al. (2020) [81], working with the same individuals as this work, demonstrate that particulate matter influenced specific morpho-anatomical traits of Q. saponaria, affecting leaf structure and potentially altering the tree’s biological functions and ecosystem services. Differences in traits like foliar perimeter, stomata characteristics, and leaf tissue thickness were observed between monitoring stations under identical irrigation conditions, highlighting the impact of urban environmental conditions. This finding indicates an effect of the local urban environmental growing condition on these morpho-anatomical traits and biomonitor characteristics. Future work needs to integrate the effects of AVOC and PM on the same individuals of Q. saponaria.

On the other hand, a study of the magnetic properties of the PM on the leaves of other urban individuals of Q. saponaria [82] shows their diverse sources, including industrial emissions and biomass burning, rather than just vehicular emissions. This work specifies the impact of vehicles on photochemical pollution, highlighting the characteristics of Q. saponaria as a biomonitor of the principal pollutants of the Metropolitan Region, O₃ and PM. The statistical analysis established that individuals of Q. saponaria adsorb a greater quantity of AVOC compared to individuals of R. pseudoacacia under similar environmental exposure conditions to the pollutants. The estimated values of Prop-equiv and OFPs were higher in Q. saponaria than in R. pseudoacacia, primarily due to the elevated concentrations of styrene and toluene. Styrene is the compound with the highest incidence in both species, followed by toluene in Q. saponaria and xylenes in R. pseudoacacia, reflecting variations in chemical reactivity based on the tree species.

The results of this study revealed a significant presence of AVOC in the leaves of urban trees, with some of these AVOC having a huge effect on human and plant health. These AVOC could originate from anthropogenic sources other than vehicular ones, for instance, industrial or domestic ones. Their presence on leaves suggests that the studied trees act as bioindicators, capturing and accumulating the atmospheric pollutants present in the urban environment. Keymeulen et al. (2001) [50] utilized Pinus sylvestris leaves as biomonitors to assess BTEX (benzene, toluene, ethylbenzene, and xylenes) levels in the air and pine leaves at three locations (roadside, gas station, and rural area) in Belgium, Hungary, and Latvia over the course of a year. Concentrations were found to be lower in the summer; however, near gas stations, the leaves exhibited significantly higher levels compared to other locations.

The results of this study show that in urban environments, Q. saponaria is a better indicator of the presence of AVOC in the gasoline range and a more efficient help to improve air quality. However, for specific industrial environments, for example, Q. saponaria may be a better indicator of the presence of ethylbenzene and m/p-xylenes and as a biomonitor for toluene. In the case of R. pseudoacacia, this species may be a better indicator of the presence of toluene and a biomonitor of ethylbenzene and 4-isopropyltoluene. In addition, both species could be used as a biomonitor for styrene.

The information obtained in this study is crucial for enhancing our understanding of the exposure of urban vegetation to specific atmospheric pollutants and their potential contribution to air quality in the Metropolitan Region of Chile. Furthermore, these results may have significant implications for the health of urban trees and their ability to perform air purification functions in urban environments.

Limitations of the Study

The leaf samples analyzed in this study were collected from young individuals of Q. saponaria and R. pseudoacacia; therefore, the reported values may not accurately reflect the contaminant capture capabilities of adult individuals. Additionally, the chemical compounds quantified in the leaf material of Q. saponaria and R. pseudoacacia primarily consist of adsorbed compounds retained in the interstices of the leaf surface. This analysis does not account for the compounds absorbed through the stomata of the leaf material; as such, absorption involves a biotransformation of the chemical compounds that falls outside the scope of this work. Consequently, the reported capacities for capturing AVOCs may be underestimated.

Future studies could expand the range of tree species investigated, examine the long-term effects of exposure to other organic pollutants, and explore absorption of them and the health of the tree species. Additionally, increasing seasonal variability in evergreen species would be beneficial, as well as assessing the additional benefits of urban forestry for climate change mitigation and public health improvement. All this information is necessary for the integration of urban forestry strategies in the city’s environmental management.

5. Conclusions

Q. saponaria and R. pseudoacacia are tree species capable of retaining AVOC in the interstices of their leaf surfaces, particularly those associated with gasoline emissions. This suggests that both species could serve as indicators and biomonitors of vehicular gasoline emissions in urban environments. The analysis of foliar material revealed significant statistical differences in total AVOC concentrations, with Q. saponaria adsorbing 38% more AVOC than R. pseudoacacia (95.2 ± 19.7 ng of C vs. 69.0 ± 29.7 ng of C). This difference was particularly pronounced for toluene, where Q. saponaria retained over 11 times more (30.3 ± 8.48 ng of C) than R. pseudoacacia (2.66 ± 1.27 ng of C). In contrast, styrene concentrations were similar, with only a 5% higher adsorption in Q. saponaria.

Regarding AVOC reactivity, Q. saponaria exhibited a 9% increase in propylene equivalents (141.1 vs. 129.6 ppbC) and a 58% greater OFP compared to R. pseudoacacia (286.3 vs. 181.6 ppbC). These results suggest that Q. saponaria is more effective in capturing AVOC and has a greater influence on O₃ formation chemistry, which marks it as a promising species for air pollution mitigation.

Given the AVOC adsorption capacity of both tree species, they could be used as biomonitors for styrene and also as a biomonitor for toluene, in the case of Q. saponaria.