Airborne reactive oxygen species (ROS) is associated with nano TiO2 concentrations in aerosolized cement particles during simulated work activities

Photocatalytic cement is self-cleaning due to the addition of titanium dioxide (TiO2) nanoparticles, which react with sunlight (UV) and produce reactive oxygen species (ROS). Construction workers using photocatalytic cement are exposed not only to cement particles that are irritants but also to nano TiO2 and UV, both carcinogens, as well as the generated ROS. Quantifying ROS generated from added nano TiO2 in photocatalytic cement is necessary to efficiently assess combined health risks. We designed and built an experimental setup to generate, under controlled environmental conditions (i.e., temperature, relative humidity, UV irradiance), both regular and photocatalytic cement aerosols. In addition, cement working activities—namely bag emptying and concrete cutting—were simulated in an exposure chamber while continuously measuring particle size distribution/concentration with a scanning mobility particle sizer (SMPS). ROS production was measured with a newly developed photonic sensing system based on a colorimetric assay. ROS production generated from the photocatalytic cement aerosol exposed to UV (3.3∙10−9 nmol/pt) was significantly higher than for regular cement aerosol, either UV-exposed (0.5∙10−9 nmol/pt) or not (1.1∙10−9 nmol/pt). Quantitatively, the level of photocatalytic activity measured for nano TiO2-containing cement aerosol was in good agreement with the one obtained with only nano TiO2 aerosol at similar experimental conditions of temperature and relative humidity (around 60%). As a consequence, we recommend that exposure reduction strategies, in addition to cement particle exposures, also consider nano TiO2 and in situ–generated ROS, in particular if the work is done in sunny environments. Graphical abstract Graphical abstract

experimental setup to generate, under controlled environmental conditions (i.e., temperature, relative humidity, UV irradiance), both regular and photocatalytic cement aerosols. In addition, cement working activities-namely bag emptying and concrete cutting-were simulated in an exposure chamber while continuously measuring particle size distribution/ concentration with a scanning mobility particle sizer (SMPS). ROS production was measured with a newly developed photonic sensing system based on a colorimetric assay. ROS production generated from the photocatalytic cement aerosol exposed to UV (3.3•10 −9 nmol/pt) was significantly higher than for regular cement aerosol, either UV-exposed (0.5•10 −9 nmol/pt) or not (1.1•10 −9 nmol/pt). Quantitatively, the level of photocatalytic activity measured for nano TiO 2 -containing cement aerosol was in good agreement with the one obtained with only nano TiO 2 aerosol at similar experimental conditions of temperature and relative humidity (around 60%). As a consequence, we recommend that exposure reduction strategies, in addition to cement particle exposures, also consider nano TiO 2 and in situ-generated ROS, in particular if the work is done in sunny environments.
Keywords Photocatalytic cement . Nano cement . Nano TiO 2 exposure . Reactive oxygen species (ROS) . ROS exposure . Health effects Introduction Nanotechnology-the study of matter in nano range from 1 to 100 nm-is widely used to improve materials' properties especially strength, weight, and insulation. In the construction sector, photocatalytic cement has been introduced for its self-cleaning properties (Lan et al. 2013;Carp et al. 2004;Banerjee et al. 2015) related to the photocatalytic activity of titanium dioxide nanoparticles (nano TiO 2 ) (Hernández-Rodríguez et al. 2019;Feng et al. 2013;Folli et al. 2010). This cement is composed of regular cement made up of fine inorganic particles such as CaO, SiO 2 , Fe 2 O 3 , and MgO (Meo 2004;Batsungnoen et al. 2019) and nano TiO 2 .
Cell toxicity associated with nano TiO 2 exposure is related to reactive oxygen species (ROS) generation, which may lead to oxidative stress, lipid peroxidation, and nucleic acid alteration (Wang and Fan 2014;Shi et al. 2013;Panieri and Santoro 2016;Liou and Storz 2010). ROS such as hydroxyl radical, superoxide anion radical, hydrogen peroxide (H 2 O 2 ), and singlet oxygen play a mechanistic role in many human diseases, including cancer (Waris and Ahsan 2006;Brieger et al. 2012), especially in the initiation and progression of multistage carcinogenesis (Waris and Ahsan 2006). Elevated ROS levels have also been associated with various inflammation-related human diseases (Alfadda and Sallam 2012).
ROS are also generated outside of the body, and has to be considered together with the endogenous ROS exposure generated through the metabolic response. Environmental ROS generation is especially relevant when airborne nano TiO 2 particulates are exposed to UV (Vernez et al. 2017). Due to its electronic energy band gap, nano TiO 2 behaves as a semi-conductor: UVexcited electrons (ē) reach the conductance band while a hole (h + ) forms at the valence energy level. The resulting ē/h + pair reacts with molecular oxygen (O 2 ) and water giving rise to a series of ROS formation. They react readily with organic materials (e.g., bacteria and mold), giving them a particularly efficient biocide property (Li et al. 2014;Lan et al. 2013;Li 2004;Chen and Poon 2009;Lee et al. 2010).
Photocatalytic cement exposure among outdoor construction workers may thus have direct exposures to ROS as secondary airborne toxicant (exogenous ROS) from UV activation of nano TiO 2 . Concentrations of exogenous ROS have not yet been assessed for these workers; consequently, potential health risks associated to this exposure are currently unknown.
Airborne ROS can be quantified using a photonic detection device that was developed at our laboratory and which relies on the formation of a colorimetric complex (Fe (III)-orange xylenol) due to the oxidation of the probe solution containing reduced iron form (Fe (II)) by ROS (Laulagnet et al. 2015). The use of multiscattering absorbance enhancement (MAE) strategy as photonic core principle for the device enabled sensitive ROS determination (Suárez et al. 2013(Suárez et al. , 2014. The main objective of the present study was to quantify amount of ROS generated from airborne cement and photocatalytic particles at constant relative humidity of about 60% under controlled conditions: i) Laboratory aerosolization with photocatalytic and regular cements equipped with a UV lamp; ii) Exposure chamber setup where two construction activities (cement bag emptying and concrete cutting) were simulated with both cement types separately.

Methods
Generation of cement aerosols Airborne particles of both photocatalytic and regular cements were generated using an aerosolization system previously described by Riediker 2015, 2016. Two grams of cement were loaded into a glass funnel and dry air blown upwards through the funnel with 2 L/min. The experimental setup is shown in Fig. 1.

Control of environmental conditions
The airborne particles produced in the funnel were transported directly into a mixing chamber by shear force and mixed with humid air originated from a nebulizer. The nebulizer flow rate was 1.5 L/min, which maintained the relative humidity at 60%. Downstream, the aerosol was driven into the exposure cylinder where they were exposed to UV radiation for 2.7 min (average residence time in the cylinder). The UV radiation light source was equipped with solar UV filters to reproduce the UV-A and UV-B spectrum. The lamp produced an irradiance intensity of 785 W/m 2 in the cylinder, corresponding to 12 folds the terrestrial irradiance. The airborne particles exiting the cylinder were captured in an impinger (25 mL) filled with FOX solution (5 mL). Temperature and relative humidity were monitored continuously during the run.
Working activities Two construction activities, cement bag emptying and concrete cutting, were simulated in an exposure chamber (10 m 3 ) with either photocatalytic or regular cement. Prior to the simulation activities, the ventilation system (80 m 3 /h) was running for 2 h in order to reduce background particles, and during simulation, the ventilation system was off. The operator simulating the construction activity wore a respirator (N100, P3, or FFP3), a chemical suit, nitrile gloves, goggles, safety shoes, and hearing protection. The bag emptying activity was performed by turning an open cement bag (25 kg) upside-down, pouring it into a plastic container (diameter × height, 60 × 40 cm), and shaking until the cement bag was empty. The concrete cutting activity was performed by using a circular saw for 10 s cutting a prepared concrete block (size 25 × 36 × 6 cm). The aerosolized cement particles were sampled in the operator's breathing zone with an impinger (25 mL) containing FOX solution (5 mL) and operating at a flow rate of 0.5 L/min. Each experimental construction activity was repeated in triplicate by a single operator, as shown in Fig. 2.
ROS analysis ROS concentration-also defined as oxidative potential-was determined using a photonic system developed by our laboratory and based on multiscattering-enhanced absorbance strategy (Laulagnet et al. 2015;Vernez et al. 2017). In brief, air samples are bubbled through an impinger filled FOX solution (5 mL), which is the reaction medium. In the presence of ROS, the Fe (II) undergoes oxidized into Fe (III) that forms a complex with orange xylenol absorbing light at 580 nm. The color change is measured via the use of a narrow emission led (580 nm) coupled to a photodetector both driven through a microcontroller board (Arduino Uno) The multiscattering regime occurring in the photonic cell due to the combination of rough aluminum cavity and inner Teflon housing enables dramatic lengthening of the optical path and improved analytical sensitivity. Statistical analysis Means and standard deviations for nanoparticle size and distribution as well as ROS concentrations were compared by two-sample t test using STATA version 15.

Results
The ROS detection system developed by our group enabled us to quantify hydrogen peroxide (H 2 O 2 ) and hydroxyl radicals (OH • ). Quantitative determination of the aerosol reactivity expressed-once normalized by total particle number concentration-in nanomoles of H 2 O 2 equivalents per particle (nmol/pt) was possible by combining accurate aerosol generation and sensitive ROS detection. The average size distribution from aerosolized photocatalytic cement in the experimental setup was around 2•10 5 pt/cm 3 , with a geometric mean diameter (GMD) of 285 nm and a geometric standard deviation (GSD) of 1.65 nm. Regular cement aerosol had a particle number concentration of 1•10 5 pt/cm 3 with GMD and GSD of 376 and 1.74 nm, respectively (Fig. 3a). The TEM images confirmed that photocatalytic cement aerosols contained agglomerates from pristine nanoparticles with primary size around 50 nm (Fig. 3b).
In complement, the aerosol ROS generation calculated in the present study indicates that the aerosolized  Riediker 2015, 2016). The main air stream was split into one leading to the SMPS for measuring particle number concentration (11-1083 nm), and a second driving the aerosol to the mixing chamber. The particles were mixed with humid air and transported into the UV-exposure cylinder (solar simulator lamp). Temperature and humidity were monitored after the air passed through the UV-cylinder. The airborne particles were captured in an impinger filled with FOX solution and the associated ROS production analyzed with the oxidative potential analyzer system (Laulagnet et al. 2015, Vernez et al. 2017 photocatalytic cement exposed to UV irradiance (3.3•10 −9 nmol/pt) is significantly more reactive in terms of produced H 2 O 2 equivalents than regular cement exposed to UV (0.5•10 −9 nmol/pt) or not (1.1•10 −9 nmol/ pt). ROS generation for non-UV-exposed cement aerosols were 1.6•10 −9 nmol/pt and 1.1•10 −9 nmol/pt for photocatalytic and regular cement, respectively (Table 1). In good agreement with a prior study (Vernez et al. 2017), the results herein obtained clearly indicate that the presence of nano TiO 2 in the photocatalytic cement do increase its chemical reactivity in terms of ROS generation prompt to act as secondary toxicants. The effect of UV irradiance on nano TiO 2 is manifest in the fact that ROS generation from photocatalytic cement doubled in the presence of UV irradiance. As expected, ROS production from photocatalytic cement exposed to UV was significantly higher than regular cement with or without UV (Fig. 4). There was no significant difference in ROS generation between regular cement exposed and not to UV light.
Simulated construction work activities performed in exposure chamber to evaluate airborne ROS levels show that for bag emptying activity, the measured ROS production was significantly greater (p value = 0.04) for photocatalytic (4.6•10 −10 nmol/pt) than for regular cement (1.5•10 −10 nmol/pt) during bag emptying (Fig. 5).
In the case of concrete cutting, no significant difference was observed between photocatalytic and regular cement, with ROS reactivities of 1.1•10 − 10 and 1.1•10 −10 nmol/pt, respectively (Table 2).

Discussion
Airborne photocatalytic cement particles are a potential source of ROS that is further enhanced in the presence of UV irradiance, and is mainly attributed to the presence of nano TiO 2 on the airborne particles (Batsungnoen et al. 2019). The photo-induced mechanism that triggers the production of ROS-mainly in the form of H 2 O 2 -at the surface of TiO 2 is wellestablished (Kakinoki et al. 2004, Ghadiry et al. 2016 and has recently been demonstrated for nano TiO 2 airborne particles in our prior study (Vernez et al. 2017). The results obtained herein clearly indicate that the presence of nano TiO 2 in the photocatalytic cement increase its chemical reactivity in terms of ROS generation, which is in good agreement with this prior study (Vernez et al. 2017).
In the absence of UV irradiance, the ROS generation associated to photocatalytic cement aerosol is not significantly different than the one obtained with regular cement, exposed or not. However, the large interval observed in the ROS production by photocatalytic aerosol without UV exposure might be attributed to the activation of nano TiO 2 by visible light during experimental measurements (Etacheri et al. 2015). Consequently, indoor construction workers using photocatalytic cement under artificial light might have greater exposures to ROS compared with workers using regular cement, although to a far lesser extent than outdoor workers.
2.2 m. Fig. 2 Experimental setup to characterize work-generated particle emissions. Two activities, bag emptying and concrete cutting, were reproduced experimentally in the exposure cabin. Workers were wearing whole body protection with personal protective equipment (PPE): dust protection cloth, rubber gloves, goggles, safety shoes, ears muff, and respirator The low reactivity observed with regular cement particles could potentially be originated from redox reactions in which transition metal in its composition-namely iron oxide (2% Fe 2 O 3 )-are prone to take part (Batsungnoen et al. 2019). While many studies have demonstrated the ability of iron oxides particles-such as Fe 2 O 3 and to a greater extent Fe 3 O 4 -to activate H 2 O 2 into highly reactive hydroxyl radical via their so-called peroxidase-like behavior (Gao et al. 2017;Pham et al. 2012), to our knowledge, the contribution of iron oxide in the generation of exogenous ROS by Portland cement particles was not yet reported.
In the case of work activities, the ROS production observed during bag emptying with photocatalytic cement was three-fold greater than the one measured with regular cement. Again, even in the absence of UV irradiance, this photocatalytic activity may be attributed to the visible light energy present in the experimental setup, though nano-TiO 2 can produce ROS also under dark conditions (Kakinoki et al. 2004). More  Distribution interval (nm) Photocatalytic cement exposed UV 3.34•10 −9 (SD = 1.32•10 −9 ) 37.35 214,482 100-930 Photocatalytic cement non-exposed UV 1.58•10 −9 (SD = 0.11•10 −9 ) 37.35 182,996 100-930 Average particle concentration (pt/cm 3 ) 198,739 Regular cement exposed UV 0.51•10 −9 (SD = 0.20•10 −9 ) 0.16 139,132 550-1000** Regular cement non-exposed UV 1.12•10 −9 (SD = 0.54•10 −9 ) 0.16 76,616 550-1000** interestingly, one can notice that in the case of concrete cutting no significant difference is shown between photocatalytic and regular concretes, while in parallel, the corresponding TiO 2 contents in the generated aerosols are relatively low (max. 2%) as shown in Table 2. The different TiO 2 content observed depending on the work activity is explained by the fact that bag emptying process favors the smaller size fraction to remain airborne (16.5% of TiO 2 detected airborne), while the larger cement powder particles will sediment rapidly.
In contrast, aerosols created from cement concrete cutting roughly reflects the initial composition of the initial cement powder in the bag (2.0% nano TiO 2 ) because the TiO 2 has become part of the cement matrix and is no longer present as individual nano TiO 2 particles. It is worthily to notice that the ROS concentration measured during bag emptying using photocatalytic cement was in the same order of magnitude than the value obtained in ROS concentration (nmol/pt) * p-value 0.01 ** p-value 0.00 *** p-value 0.00 Photocatalytic cement exposed UV Photocatalytic cement non-exposed UV Regular cement non-exposed UV Regular cement exposed UV * ** *** prior work with pure nano TiO 2 once normalized by TiO 2 content (Vernez et al. 2017). Finally, health effects related to airborne nano TiO 2 exposure in photocatalytic cement should integrate its reactivity-in the presence of environmental UV/vis irradiance-by considering the associated ROS products as secondary airborne toxicants. In terms of toxic effects, ROS are associated to various metabolic/ pathological paths such as oxidative stress, inflammation, genotoxicity, cytotoxicity, DNA damage, and cancer (Li et al. 2014;Brieger et al. 2012;Scherz-Shouval and Elazar 2011;Jaeger et al. 2012;Yin et al. 2012;Jaeger et al. 2012;Wang and Fan 2014).

Conclusion
The combination of an efficient aerosol generation setup coupled with a solar simulation lamp and a sensitive photonic detection device made it possible to assess the production of ROS by photocatalytic and regular cement aerosols. As expected, the presence of nano TiO 2 in photocatalytic cement has a strong impact on the ability of the corresponding aerosol to produce exogenous airborne ROS in the presence of UV light. Moreover, the level of ROS generated during work activities was found to be linked to the amount of airborne nano TiO 2 present in the cement aerosol. Thus, concrete cutting activities appear to be considerably less problematic in terms of ROS production than bag emptying for which the nano TiO 2 content in the aerosol reaches 16%. Considering the photoreactivity of aerosolized photocatalytic cement under UV irradiance and the high content of airborne nano TiO 2 generated during bag emptying, worker protection procedures should not only consider nano TiO 2 exposure but also its ability to produce ROS as secondary airborne potential toxicants. Providing the specific reactivity of its aerosol under environmental conditions, photocatalytic cement should not only be considered a novel promising material but also a potential new hazard in construction sites.
Funding information This study was supported by the Center for Primary Care and Public Health (Unisanté), Department of Occupational and Environmental Health (formerly known as Institute for Work and Health, IST), Switzerland, together with the Royal Thai Government and the Ministry of Science and Technology, Thailand.

Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.