SARS-CoV-2 removal with a polyurethane foam composite

Abstract

The pandemic of COVID-19 (SARS-CoV-2 disease) has been causing unprecedented health and economic impacts, alerting the world to the importance of basic sanitation and existing social inequalities. The risk of the spread and appearance of new diseases highlights the need for the removal of these pathogens through efficient techniques and materials. This study aimed to develop a polyurethane (PU) biofoam filled with dregs waste (leftover from the pulp and paper industry) for removal SARS-CoV-2 from the water. The biofoam was prepared by the free expansion method with the incorporation of 5wt% of dregs as a filler. For the removal assays, the all materials and its isolated phases were incubated for 24 h with an inactivated SARS-CoV-2 viral suspension. Then, the RNA was extracted and the viral load was quantified using the quantitative reverse transcription (RT-qPCR) technique. The biofoam (polyurethane/dregs) reached a great removal percentage of 91.55%, whereas the isolated dregs waste was 99.03%, commercial activated carbon was 99.64%, commercial activated carbon/polyurethane was 99.30%, and neat PU foam reached was 99.96% for this same property and without statistical difference. Those new materials endowed with low cost and high removal efficiency of SARS-CoV-2 as alternatives to conventional adsorbents.

Graphical abstract

Introduction

The COVID-19 pandemic declared by the World Health Organization (WHO) on March 11, 2020 (WHO 2020) has spread rapidly around the world causing negative epidemiological, social, economic, cultural, and political impacts. SARS-CoV-2 is a positive-sense single-stranded RNA virus belonging to the Coronaviridae family (Polo et al. 2020). This respiratory syndrome may yield certain symptoms, such as fever, cough, shortness of breath, damage to the respiratory, hepatic, neurological systems, and even death in some cases (Khan and Yadav 2020; Wong et al. 2019).

This disease is transmitted by droplets from breathing, coughing, sneezing and direct touching (La Rosa et al. 2020), which demanded behavioral changes related to social isolation and closure of institutions for controlling the dissemination of this disease (Nghiem et al. 2020). This virus has high time of incubation and elimination and, besides of that, some infected people remain asymptomatic, which contributes for the current global pandemic scenario (Hart and Halden 2020).

This pandemic also raised an alarm on the restrict access to sanitation and the social inequalities that exist in the whole world (Daughton 2020). The current governmental institutions should take into account the eminent risk of spreading of new diseases since COVID-19 is considered the most impactful infectious disease after the Spanish flu pandemic from 1918 (Polo et al. 2020; Hart and Halden 2020). This is the third outbreak due to a viral zoonotic disease in the last two decades, succeeding SARS from 2002 and MERS from 2012 (Nghiem et al. 2020; Tortora et al. 2012).

Some authors also discussed the contamination and retention of SARS-CoV-2 in waters, as well as the potential of this contaminated effluents to infect people (Amoah et al. 2020; Ahmed et al. 2020; Bhowmick et al. 2020; Mandal et al. 2020; Xiao et al. 2020). Countries that lack efficient pathogen removal systems may pose a high risk of contamination (Bhowmick et al. 2020) and the pathway and rate of virus breakdown may vary depending on piping, septic tank employment, and watershed size (Polo et al. 2020). The development of materials for disease protection, disease detection, and water treatment is critical to efforts toward stopping the pandemic of COVID-19 since to control emerging pathogens in wastewater may mitigate the risk against public health (Tang et al. 2020; Lahrich et al. 2021).

Polyurethane (PU) foams are important engineering materials for acoustic and thermal insulation, automotive industry, and household and marine applications (Akindoyoet al. 2016; Cinelli et al. 2013; Delucis et al. 2018; Tan et al. 2011). Distinct fillers incorporated in polymer foams have been yielding cellular materials with low environmental impact and low cost (Brito et al. 2011; Tan et al. 2011), as well as increased performance for several applications (Barreto et al. 2016; Brito et al. 2011; Cinelli et al. 2013).

Brazil was the world’s second largest producer of cellulosic pulp in 2018 (IBÁ 2019) with intense participation in technological development and industrial facilities with high productivity (Moura et al. 2018). This basic raw material of paper is majorly produced following the Kraft process, which aims to dissolve the lignin that connects the cellulosic fibers using sodium salt solutions, although this industrial process also generates woody residues, black liquor, ashes and sludges from effluent treatment plant, and residues from chemical recovery, such as dregs, grits, and paper sludge (Alves et al. 2015; Borges et al. 2016).

There are several studies on the application of wastes leftover from the pulp and paper industry in order to reduce their negative environmental impacts, including construction materials (Marques et al. 2014; Mymrin et al. 2016), substrate for soils (Toledo et al. 2015), sanitation applications (Oliveira et al. 2017; Orlandi et al. 2017), and removal of contaminants (Farage et al. 2020). PU bio-foams filled with this type of waste could be applied as a new absorbent for the removal of pathogens from the water since PU foams were already applied as adsorbent supports. The objective of this work was to investigate the application of rigid polyurethane biofoams incorporated with dregs, commercial activated carbon (CAC) and its isolated phases as adsorbents for the removal of SARS-CoV-2 from contaminated water.

Experimental

Preparation of raw materials

Green liquor dregs wastes were supplied by CMPC located in Guaiba/Brazil. This residue and commercial activated carbon (CAC) (PA, Dinâmica) were dried at 50 °C and sieved (100-mesh screen; aperture of 150 µm). Neat PU, CAC/PU, and dregs/PU were prepared by the free expansion method using two mixture components (A and B) at a 1:1 NCO/OH ratio and 5% filler content (Delucis et al. 2018). Component A consisted of castor oil (hydroxyl content of 160 mg KOH·g⁻¹), glycerin P.A., dregs/CAC, chain extender (polyethylene glycol), surfactant (Tegostab B804), and distilled water, which was homogenized for 60 s at 1000 rpm under mechanical stirring and was then left to degas for 120 s. Component B is catalyst (Tegoamin DMEA) and a polymeric MDI (diphenylmethane diisocyanate), which was added to the component A and then stirred for 20 s under mechanical agitation. The final mixture was poured into an open mold and left to rise for 24 h. The solid foam was cured at 60 °C for 2 h in an oven and post-cured at 65% relative humidity and 20 °C for 2 weeks, as recommended by the literature (Delucis et al. 2018).

Scanning electron microscopy

Surface morphologies of the different materials were obtained by scanning electron microscopy (SEM) (JEOL, JSM 6610LV, Japan). The working voltage was 15 kV and the magnification of 100 × .

X–ray diffraction

X–ray diffraction (XRD) patterns were obtained using a diffractometer (Brunker D–8, Germany), provided with a diffracted beam monochromator and Ni filtered CuKα radiation (λ = 1.5406 Å). The voltage was of 40 kV and the intensity of 40 mA. The 2θ angle was scanned between 10° and 60°, and the counting time was of 1.0 s at each angle step (0.02°).

Fourier-transform infrared spectroscopy

Chemical groups were obtained with Fourier-transform infrared spectroscopy (FT-IR) using IRPrestige-21 (Shimadzu, Japan) scanning from 500 to 4000 cm⁻¹, 32 scans, transmittance mode, and resolution of 4 cm⁻¹.

Point of zero charge

Point of zero charge (PZC) were obtained using the 24-h agitation contact at 50 rpm in initial pH solutions that varied from 1 to 12. The PZC was obtained after plotting the ΔpH (pH final – pH initial) versus initial pH. This methodology was adapted from that described by Farage et al. (2020).

SARS-CoV-2 inactivated

An inactivated SARS-CoV-2 virus used as a positive control and comes from a clinical isolated in Vero-E6 cell culture (SARS.COV-2/SP02/human2020/Br, GenBank accession number MT126808.1). This virus was kindly provided by Prof. Dr. Edison Luiz Durigon from Department of Microbiology, Institute of Biomedical Sciences, University of São Paulo (USP), Brazil (Dorlass et al. 2020).

Removal of SARS-CoV-2 from the water

A total of 10 mg of each adsorbent were properly dried at 37 °C for 2 h. Afterwards, the adsorbent was transferred to a microtube containing 1.5 mL of ultrapure water (free of all RNAse enzymes) and 150 µL of the inactivated SARS-CoV-2 viral suspension (2.5 × 10⁶ copies/mL) was then added, which was followed by incubation with shaking at 200 rpm and 28 °C for 24 h. Subsequently, both supernatant and adsorbent were removed and placed into another microtube, and the viral RNA was then extracted. The virus adsorption was calculated and presented as described by Demarco et al. (2022). The viral load removed was calculated using the following Eq. 1:

$$\mathrm{Viral}\mathrm{load}\mathrm{removal}=〚\mathrm{Viral}\mathrm{load}〛\_(\mathrm{supern})-〚\mathrm{Viral}\mathrm{load}〛\_\mathrm{mat}$$ 1

where viral load removal is expressed in copies mL⁻¹,Viral load supern refers to the viral load in supernatant (copies mL⁻¹), and Viral load mat refers to viral load in material(copies mL⁻¹).

The CT values are inverse to viral load content, and it is an indirect method for detection of copy number of viral RNA (Rao et al. 2020). Table 1 demonstrates the equivalence for transforming CT values in viral load.

Table 1.

CT values and correspondent viral copies mL⁻¹

CT	Viral copies mL−1	Ct	Viral copies mL−1
15	2,500,000	28.5	228.75
16	1,250,000	29	152.5
17	625,000	29.5	114.37
18	312,500	30	76.25
19	156,250	30.5	57.18
19.5	117,187.5	31	38.12
20	78,125	31.5	28.591
20.5	55,593	32	19.06
21	39,062	32.5	12.296
21.5	29,296	33	9.531
22	19,531	33.5	7.145
22.5	14,647	34	4.765
23	9765	34.5	3.57
23.5	7323	35	2.38
24	4882	35.5	1.78
24.5	3661	36	1.19
25	2441	36.5	0.88
25.5	1830	37	0.59
26	1220	37.5	0.438
26.5	915	38	0.29
27	610	38.5	0.222
27.5	457.5	39	0.148
28	305	39.5	0.111
		40	0.074

RNA extraction

The RNA was extracted from both supernatant and studied adsorbents using a MagMax™ Core Nucleic Acid Purification kit (Thermo Fisher Scientific, Waltham, MA, USA). The extracted RNA was quantified by Nano Drop® (Thermo Scientific, Waltham, MA, USA). A concentration of approximately 10 ng of RNA was used to perform the RT-qPCR detection.

qRT-PCR

The primer and probe used in PCR reactions was designed according to the sequences published by the Centers for Disease Control and Prevention (CDC 2020). Briefly, a reaction of 25 μL of final volume was used, with the following volumes added to the 1 × concentrated master mix: 5 μL of sample RNA, 12.5 μL of 2 × reaction buffer, 1 μL of Superscript™ III One-Step with Platinum™ Taq DNA Polymerase (Invitrogen, Darmstadt, Germany), 0.4 mM of each dNTP, 0.4 μL of a 50 mM MgSO4 solution (Invitrogen), 1 μg of non-acetylated bovine albumin (Roche), 10 μM of each primer 2019-nCoVN1-F2019-nCoV N1 (5′GACCCCAAAATCAGCGAAAT3 ′), 2019-nCoVN1-R2019-nCoV N1 (5′TCTGGTTACTGCCAGTTGAATCTG3 ′), 2019-nCoVN1-P2019-nCoV N1 probe (5′-FAM – ACCCCGCATTACGTTTGGTGGACC– BBQ 3′), and DEPC water. The reaction occurred in StepOne™ Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA, USA) in the following cycling: 55 °C for 10 min for reverse transcription, followed by 95 °C for 3 min and 40 cycles of 95 °C for 15 s, 58 °C for 30 s.

Statistical analysis

Data were expressed as mean ± standard deviation for duplicates for each experimental point. Data were analyzed by using one-way analysis of variance (ANOVA) followed by Bonferroni’s multiple comparison tests adjusted for a significance level of 5%.

Results

Figure 1 shows SEM images of the studied adsorbents. The dregs and CAC, which were analyzed as received, seems to be composed of rough particles and some aggregates, whereas the neat PU foam and CAC/PU, in turn, presented rounded polymer cells with about 100 μm in diameter. Lastly, dregs/PU was mostly composed of irregular shaped cells than CAC/PU.

Fig. 1.

SEM images for dregs (a), CAC (b), dregs/PU foam (c), CAC/PU (d), and neat PU foam (e)

Figure 2 shows the XRD diffractograms of the studied adsorbents. The dregs (shown in Fig. 2a) showed crystallinity with prominent peaks in 2θ angle of 26°, 28°, 35°, 40°, 45°, 47°, and 50° that can be attributed to its high ashes content probably derived from thermally decomposed woods. Compared to the dregs, this crystallinity is smaller for the filled foam. The diffractogram shows a crystalline peak at a 2θ angle of 28º associated with the dregs waste. The CAC presented 2θ angle of 20°, 26°, 36°, 50°, 59°, and 67° predominantly crystalline attributed to the presence of ash resulting from pyrolysis and activating agent outs. With the addition of CAC to polyurethane, there is a suppression of the halo existing in 2θ, probably due to the presence of activated carbon, but maintaining the strongly amorphous character of the material.

Fig. 2.

XRD for dregs (a), CAC (b), dregs/PU foam (c), CAC/PU (d), and neat PU foam (e)

The spectrum ascribed to the dregs (Fig. 3a) present prominent bands at 1390 cm⁻¹, 869 cm⁻¹, and 711 cm⁻¹. Based on the spectra of neat and filled PU foams (Fig. 3b, c, d), it is possible to observe almost the same prominent bands at 3310 cm⁻¹, 2837 cm⁻¹, 2274 cm⁻¹, 1708 cm⁻¹, 1513 cm⁻¹, 1209 cm⁻¹, and 1042 cm⁻¹. For CAC the bands of 2158 cm⁻¹, 2029 cm⁻¹, 1978 cm⁻¹, and 1637 cm⁻¹ are characteristic. There were no prominent bands ascribed to the dregs and CAC in the infrared of the filled foam, although both the SEM and XRD results confirmed the presence of this residue in the biofoam structure.

Fig. 3.

Infrared spectra for dregs (a), CAC (b), dregs/PU foam (c), CAC/PU (d), and neat PU foam (e)

Pzc of dregs was 8.40, CAC was 6.60, neat PU was 6.46, CAC/PU was 6.64, and PUD was 7.14. Associate an influence of the dregs waste in the increase in foam pzc, whereas, the pzc of the material proposed in this study is in a relatively intermediate value between the waste and the pure foam. Both pcz for CAC and neat PU were similar, resulting in the same range for CAC/PU (Fig. 4).

Fig. 4.

Point zero change for (a), CAC (b), dregs/PU foam (c), CAC/PU (d), and neat PU foam (e)

Table 2 indicates that the supernatant, neat PU, CAC/PU, and dregs did not differ from each other in terms of CT values. The CT of all materials were similar, for neat PU 31.93 ± 2.82, for dregs 28 ± 0.98, CAC for 32.68 ± 5.99, for CAC/PU 23.12 ± 0.83, and for dregs/PU 29.72 ± 0.40.

Table 2.

Cycle threshold (CT), viral load (copies mL⁻¹), and removal properties obtained after 24 h of incubation

	PU	dregs	dregs/PU	CAC/PU	CAC
Supernatant CT	20.74 ± 1.58	21.39 ± 0.38	26.11 ± 0.77	23.12 ± 0.83	24.73 ± 0.69
viral load in supernatant (copies mL−1)	47.65 × 103	31.35 × 103	1.14 × 104	9.15 × 103	3.09 × 103
material CT	31.93 × 2.82	28 ± 0.98	29.72 ± 0.40	30.32 ± 1.11	32.68 ± 5.99
viral load in material (copies mL−1)	0.020 × 103	0.310 × 103	0.097 × 103	0.064 × 103	0.011 × 103
viral load removed (copies mL−1)	47.64 × 103	31.04 × 103	1.05 × 103	9.09 × 103	3.09 × 103
viral load removed (copies mL g−1)	4.76 × 106	3.10 × 106	0.10 × 106	0.91 × 106	0.31 × 106

Values are mean ± standard deviation. Control represents free SARS-CoV-2 viral particles, C_T = 14.85 ± 0.96 and viral load in control (copies mL⁻¹) = 2.5 × 10⁶ ± 0.11 × 10⁶. Viral load removed (copies mL⁻¹) = [viral load in supernatant − viral load in material]

The RT-PCR technique is based on the detection of amplification cycles and is an indirect method for determining viral RNA copies, wherein the CT values are inversely proportional to the viral load (Bustin et al. 2005; Mutesa et al. 2021). In a RT-PCR analysis, the number of viral DNA molecules is doubled at each cycle (Mutesa et al. 2021). A reference number of 2.5 × 10⁶ viral copies per mL was considered for 15 cycles. In this sense, the viral loads (VC) of adsorbent, supernatant and that removed were not equals.

The CV removed per gram of adsorbent was 4.76 × 10⁶ for the neat PU, 0.91 × 10⁶ for CAC/PU, 0.31 × 10⁶ for CAC, whereas dregs and dregs/PU reached 3.10 × 10⁶ and 0.10 × 10⁶, respectively. Besides of that, an outstanding percentage SARS-CoV-2 removal of 99.03% was reached for the dregs, whereas this property for the neat PU was 99.96%, 99.64% for CAC, filled foams were 91.55% for dregs/PU and 99.3% for CAC/PU (Fig. 5), when compared to the control of free SARS-CoV-2 in water. No significant statistical differences were observed among the materials tested.

Fig. 5.

Percentage SARS-CoV-2 removal (%) of with PU, dregs, dregs/PU, CAC/PU, and CAC. A triple asterisk represents values with a significant difference in relation to the Control with p˂0.05. Control represents free SARS-CoV-2 viral particles in water without the materials studied

Discussion

The irregularly shaped particles sometimes forming agglomerates founded for the studied dregs is typical of this industrial solid waste (Mymrin et al. 2016). Diffractogram peaks at 2θ angles between 25º and 30° indicates the presence of a crystalline fraction of calcite (CaCO₃) and other minor minerals, such as perovskite (Ca₄Ti₄O₁₂), dolomite (CaMg(CO₃)₂), quartz (SiO₂), and manganite (Mn₄O₈H₄) (Mymrin et al. 2016; JIA et al. 2019; Quina and Pinheiro 2020). The infrared spectrum obtained for the dregs corroborates those reported previous studies, in which intense bands near 1428 cm⁻¹, 874 cm⁻¹, and 710 cm⁻¹ indicate the presence of calcium oxides and salts, and magnesium carbonate (Almeida et al. 2007; Matias 2012). For CAC, MEV indicates abundant pore structure and smooth surface, 2θ angles of 20° and 25° represent the presence of carbon and calcium carbonate (Shu et al. 2017), and the bands near 1637 cm⁻¹ indicate the aromatic elongation of C = C (Isik-Gulsac 2016).

Both neat and filled foams presented a cellular structured marked by numbers open cells, which is normal for PU foams applied as perimeter upstand insulation (Cinelli et al. 2013; Delucis et al. 2018). Also, the incorporation of dregs fillers in PU foams may increase its number of open cells due to some disrupted edges when the filler attaches itself to the polymer cell wall (Delucis et al. 2018). None foams show an organized crystalline phase, which is confirmed by the absence of clear diffraction peaks (Almeida et al. 2020; Schio et al. 2019).

Those prominent bands founded at 3310 cm⁻¹ and 1513 cm⁻¹ are related to the presence of the N–H bonds from urethane groups ((–NH–(C = O)–O–) belonging to the PU cell wall (Kumari et al. 2016; Schio et al. 2019; Delucis et al. 2018). That band at 2837 cm⁻¹ is associated with aliphatic groups (Santos et al. 2017; Kumari et al. 2016), whereas the band at 2274 cm⁻¹ represents vibrational modes of N = C = O bonds from isocyanate groups (Santos et al. 2017; Kumari et al. 2016). The latter band may also represent unreacted NCO groups (Schio et al. 2019), which indicates that the dregs imparted a negative effect to the polymerization process, probably related to a steric hindrance mechanism (Delucis et al. 2018).

Pzc measures the potential of the absorbent surface to become positive or negative and is the value at which the net charge of the adsorbent becomes zero. Farage et al. (2020) founded a pzc of 9.75 for a certain dregs waste, which was attributed to its high calcium content. Almeida et al. (2018) reported a pzc of 6.05 for their PU foam, which was related to the generation of surface charges from protonation/deprotonation of N–H groups from urethane bonds, which were detected here by the FT-IR analysis.

Viruses are negatively charged colloidal particles with almost neutral pH values and with the ability to adsorb certain substance onto their surfaces (Lahrich et al. 2021). Some virus removal processes in water treatment occurs due to adsorption between the virus and the suspended solid particles (Bitton 1975; Gundy et al. 2009). This fact may be associated with the higher removal capacity attributed to the dregs in a comparison with the studied PU foams since the dregs is composed of thin particles endowed with a high surface area, onto which dissolved viruses may be adsorbed.

De Wit et al. (2015) reported viruses (called as nanobacteria) adsorbed onto calcite grains in natural environments. This study corroborates the interaction between viruses and minerals by precipitation processes. Carter et al. (2021) also cited those minerals can have antiviral activities and bind themselves to virus particles.

The removal mechanism is probably related to the active sites from the adsorbents and the S protein of the virus, which is responsible for binding with the host cell (UZUNIAN 2020). SARS-CoV-2 may be direct adsorbed by electrostatic interactions with oppositely charged surfaces due to –NH₂, –NH₃⁺, –COOH, and –COO⁻ groups from its amino acids. This could not occur for the studied adsorbents due to their negative charge. However, this virus may have a positive charge below its isoelectric point, which could allow the protonation of functional groups and formation of hydrogen bonds (Joonaki et al. 2020). According to Fuhs et al. (1985), electrostatic bonding between mineral surfaces and viruses may occur by Van Der Waals forces.

Carrero et al. (2011) cited that the particle size of milled PU foams may hinder an adhesion mechanism with contaminants since the size defines the “path” that the contaminant has to travel until the binding site. But irregular cells of the dregs/PU may have negatively inferred in the ability to remove which compare to neat PU and CAC/PU.

Even so, both the studied foams showed good results for the removal of SARS-CoV-2 and the dregs insertion yielded an improved removal. A PU-based antimicrobial material developed by Park et al. (2013) also showed antiviral activity with removal capacity of 2.2 ± 0.3 × 10⁴ mL⁻¹ and 1.72.2 ± 0.4 × 10⁵ mL⁻¹ for Influenza and Poliovirus, respectively.

Regarding the coronavirus removal in room temperature water, 10 days are required for a 99.9% CV removal (Gundy et al. 2009). Haramoto et al. (2020) reported 2.4 × 10³ L⁻¹ CV of SARS-CoV-2 studding a treated wastewater from Japan. Peccie et al. (2020) announced viral RNA copies that varied from 1.7 × 10³ to 4.6 × 10⁵ mL⁻¹ studding SARS-CoV-2 in primary sludge. Hart and Halden (2020) estimated the presence of 0.15 to 141.5 × 10⁶ L⁻¹ of SARS-CoV-2 viral genomes studding municipal wastewaters from North America and Europe. The CV removal capabilities found in this study would be sufficient to largely eliminate the viral concentrations reported in the literature.

In pandemic scenarios the control of the water cycle is extremely important to prevent the spread of viruses (Wigginten and Elleberg 2015), especially in the current SARS-CoV-2 outbreak. Water decontamination practices must be encouraged for an effective removal of microorganisms, viruses, and other contaminants (Amoah et al. 2020; La Rosa et al. 2020).

Conclusions

Green liquor dregs waste and commercial activated carbon were successfully incorporated into a rigid polyurethane foam and both foam composite and its isolated phases were characterized for chemical and morphological features. All materials were also tested for SARS-CoV-2 removal. Therefore, the surface of this inorganic filler, which is mainly composed of calcite (CaCO₃), probably chemically bonded itself to the virus. Further studies may address increased filler contents and field tests in contaminated areas.

Author contribution

Guilherme Pereira Schoeler: material preparation, data collection, and analysis were performed. Thays França Afonso: material preparation. Carolina Faccio Demarco: material preparation. Victor dos Santos Barboza: material preparation. Tito Roberto Sant’annaCadaval: analysis were performed. Andrei ValerãoIgansi: analysis were performed. Marcos Alexandre Gelesky: analysis were performed. Janice Luehring Giongo: material preparation. Rodrigo de Almeida Vaucher: material preparation, data collection, and analysis were performed. Rafael de Avila Delucis: material preparation. Robson Andreazza: material preparation, data collection, and analysis were performed. All authors read and approved the final manuscript.

Funding

This study was partly supported by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance code 001, by the CNPq (National Council for Scientific and Technological Development) and FAPERGS (Research Support Foundation of the State of Rio Grande do Sul).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethical approval

Not applicable.

Consent to participate

Not applicable.

Consent to publish

Not applicable.

Conflict of interest

The authors declare no competing interests.