Comparative toxicity and environmental impact assessments of sonochemically-synthesized CuO and Zn-doped CuO nanoparticles using zebrafish and LCA tools

Abstract

Nanomaterials (NMs), including nanoparticles (NPs), offer promising potential in achieving the European Commission’s Green Deal goals of climate-neutral, zero-pollution and circular economy. Metal oxide NPs display antimicrobial properties, with efficacy also towards antimicrobial-resistant bacteria. Nevertheless, the increasing manufacture, use and unintended release of NMs particularly in aquatic compartments, raises concerns about their environmental sustainability and safety towards non-target organisms. Within the Safe and Sustainable by Design framework, this study compares toxicity and environmental impacts of sonochemically synthesized water-based CuO and Zn-doped CuO NPs. Zebrafish embryos were exploited in a high-throughput developmental and behavioral screening to investigate nanosafety. The Fish Embryo acute Toxicity test was used to assess the NPs aquatic toxicity potential, while behaviour was addressed by tracking embryos activity. The Life Cycle Assessment (LCA) methodology was implemented through the OpenLCA software to evaluate the environmental footprint of the NPs synthesis. Our findings showed no significative lethality at the tested concentrations (0.01–100 mg/L) (LC_{50 wCuO} >  > 100 mg/L), with the exception of ZnCuO NPs 100 mg/L (LC_{50 ZnCuO} = 123 mg/L). Sub-lethality occurred as delayed hatching, partially recovered by Zn-doping, and embryo development. LCA highlighted the dominant role of electricity (which represented 47 to 98% of the total impacts) and copper acetate (37–94%) consumption in the environmental impacts of the NPs synthesis, emphasizing the importance of optimizing energy and chemical use to minimize environmental burden. This research supports the safe and sustainable design of nano-enabled antimicrobials and underscores the need for an approach comprehensive of both risk assessment and LCA in nanotechnology development.

Supplementary Information

The online version contains supplementary material available at 10.1186/s11671-025-04225-7.

Introduction

Nanomaterials (NMs), such as nanoparticles (NPs) and nano-enabled products (NEPs), are pivotal to achieving the European Commission’s Green Deal goals of a climate-neutral, zero-pollution, sustainable, and circular economy [1]. Their rapid development has enhanced functionalities compared to bulk materials, making them valuable in medicine, food, and environmental cleanup [2, 3]. Furthermore, NMs display proven efficacy as antimicrobial agents against antimicrobial-resistant (AMR) bacteria [4, 5]. Metal and metal oxide NPs (e.g., Ag, CuO, ZnO) exhibit strong bactericidal properties towards different strains of bacteria [6]. Particularly, copper oxide (CuO) and its compounds have been historically employed as disinfectant agents to contrast bacteria, but also fungi and viruses, both as surface and particle agent [7]. Zinc oxide (ZnO) is also known for its biocidal properties towards a broad spectrum of bacteria [8, 9].

One of the major factors contributing to the antimicrobial potential of MeO NPs is the release and dissolution of metal ions in the environment, or the exposure medium, which seems to have a direct impact on cell membrane [10, 11]. Furthermore, following interaction with aqueous solutions, MeO NPs produce reactive oxygen species (ROS) [8] that are responsible for killing bacteria as a consequence of oxidative stress and damage, altered antioxidant response and impairment of bacteria morphology and structure [4]. Unfortunately, dissolution properties and the creation of ROS by MeO NPs play a major role in hazard potential also to non-target organisms [12, 13].

The doping of MeO NPs with other metals has been shown to modulate the physicochemical properties of these NMs, influencing their toxicity. [14] demonstrated that doping CuO NPs with Fe alters their dissolution kinetics in aqueous biological environments, resulting in safer NMs that have a reduced hazardous impact by decreasing copper ion release. In contrast, other studies have shown that doping CuO with Mg and Zn increases the production of ROS, leading to enhanced antibacterial activity due to a higher number of surface structural defects [6, 15].

Despite the significant potential of CuO and Zn doped CuO NPs as antimicrobial agents, their increased production and use in NEPs, such as antimicrobial water filtration membranes, may lead to improper disposal in aquatic environments, posing risks to non-target species [12]. Studies have demonstrated the presence of engineered and metal-based NPs in aquatic environment, particularly near wastewater treatment facilities [3, 16, 17], and highlighted adverse effects associated with these NPs in both in vitro and in vivo models.

Recognizing the significance of ecotoxicological assessments of NMs, the Safe and Sustainable by Design (SSbD) framework, endorsed by the European Commission’s Green Deal, emphasizes the need to integrate safety considerations with environmental sustainability and functionality at the early stages of material development [1]. This necessitates an assessment of both the safety and sustainability of nanomaterials throughout the innovation process, ensuring that new technologies meet performance requirements while minimizing risks to human and environmental health across their entire life cycle [18].

Indeed, studies have highlighted that the synthesis phase of CuO NPs often requires high temperatures and large energy amounts [19], which may lead to significant environmental burdens, such as global warming potential and resource depletion. Additionally, existing synthesis methods may utilize toxic reagents, posing risks to both ecological and human health [20].

In line with the concept of SSbD, this study aims to evaluate the nanosafety and sustainability of two novel antimicrobial CuO based NPs, produced via a one-step sonochemical synthesis, whose antibacterial efficacy against multidrug-resistant bacteria has been previously demonstrated [6]. In particular, to compare design variables of the physicochemical structure and the synthesis process, we considered water-based CuO (wCuO) versus ethanol-based Zn-doped CuO (ZnCuO) NPs. Considering the potential for NPs release during their lifecycle into aquatic environments, we have chosen to assess the toxicity of these NPs on zebrafish embryos using the Fish Embryo Acute Toxicity (FET) test [21]. Zebrafish embryos serve as a powerful in vivo model for ecotoxicity studies, with the advantage to bridge the gap between in vitro assays and mammalian studies while adhering to ethical research practices through the 3Rs principles [22]. Furthermore, the first steps of the NPs life cycle were addressed through the Life Cycle Assessment (LCA) standardized methodology [23, 24] following a cradle-to-gate approach to provide the sustainability profile of the synthesis processes.

By combining FET test outcomes and other sublethal endpoints with LCA findings, this work introduces an innovative approach to characterize the environmental hazards associated with the sonochemical synthesis of CuO NPs and identify safer and more sustainable alternatives for developing nanoCuO-based antimicrobials. This innovative approach integrates sonochemical synthesis with LCA within an ecotoxicology framework, focusing on water-based methodologies in contrast to traditional ethanol-based techniques [6]. To our knowledge, this dual approach has not been sufficiently explored to date, also because of the lack of nano-specific impact categories that hinders the full application of LCA to NMs field, highlighting the need for further research in this area. This comprehensive approach enhances understanding of the ecotoxicological and environmental implications of NP production, aligning with the objectives of the EU Green Deal. Furthermore, it lays the foundation for an integrated evaluation of the safety and sustainability of the novel antimicrobial wCuO and ZnCuO NPs, as well as NMs in general.

Materials and methods

Chemicals

All analytical-grade reagents, 3-amino-benzoic acid ethyl ester (MS222), FET salts, microscopy reagents were purchased by Merck KGaA (Darmstadt, Germany). Instant Ocean salt was purchased from Aquarium Systems (Sarrebourg, France). The starting materials for synthesis of metal oxides are corresponding salts of acetates that were hydrolysed in an alkaline environment for formation of CuO and ZnCuO NPs under ultrasound irradiation.

Nanoparticles synthesis

The NPs were synthesized by sonochemical techniques in Bar-Ilan University, Center for Advanced Materials and Nanotechnology (Israel) in the framework of the EU projects PROTECT and AMROCE. Sonochemical synthesis was chosen over other methods since was proven as effective, cost-effective, up-scalable and provide a stable coating which can withstands washing cycles [25]. As already reported in [26], the synthesis mainly consisted in the dissolution and sonication of copper and zinc acetate as precursors in aqueous solution. Additionally, an aqueous solution of ammonium hydroxide (28–30%) was added to adjust the pH to 8. CuO NPs were prepared by dissolving 0.6 g of copper acetate in the 300 mL aqueous solution. The solutions were sonicated by probe sonicator (20 kHz, 750 W, 35% amplitude). When the temperature reached 60 °C, an aqueous solution of ammonia was added to adjust the pH to 8. The sonication was conducted for 30 min. The resultant solution was centrifuged, washed with water and dried at 60 °C overnight. A detailed procedure for the synthesis and characterization of ZnCuO is available in [6]. Ethanol 90% is also required for the synthesis of ZnCuO NPs with a 9:1 volume ratio (ethanol:deionized water). At the next stage, the powders were submitted to the toxicological analyses.

Life cycle assessment

The analysis of the environmental impacts was performed using the standardized LCA methodology [23, 24], which includes the phases of (a) goal and scope definition, (b) inventory analysis, (c) impact assessment (LCIA) and (d) interpretation. The methodological phases and LCA study details are briefly described below.

(a) The goal and scope of this work was to assess in a “cradle to gate” perspective the absolute and comparative environmental impacts associated to the lab scale production of wCuO NPs and ZnCuO NPs purposed for antimicrobial functionality for biomedical and aquaculture applications, and obtained through two different sonochemical syntheses processes.

The use and end-of-life stages were not included in the study, as they were beyond the scope of the PROTECT and AMROCE projects. The reference Technical Unit was associated to the amount of 5.5 g of NPs obtainable by the two synthesis routes.

(b) Within the inventory phase, information and primary data about input and output flows (in terms of material, energy, waste) on the NPs’ synthesis processes were collected within the scope of PROTECT and AMROCE projects. Secondary data related to upstream processes associated to raw materials and energy implied in the synthesis process were derived by the Ecoinvent 3.7 database.

Details about the synthesis of the nanoparticles are reported in Sect. 2.2. As ethanol can be entirely recovered after the NPs’ synthesis through a distillation process, its input was omitted in the impacts’ calculation, while the energy required for its recovering through distillation process (0.185 kWh/L) was considered.

The inventory data used to model the synthesis processes are reported in Table 1.

Table 1.

Inventory data for the synthesis processes of 5.5 g of NPs

Input	wCuO NPs	ZnCuO NPs
Copper acetate (g)	22.0	21.0
Zinc acetate (g)	/	7
Deionized water (L)	10.0	10.0
Ammonium hydroxide (28–30%, aqueous solution) (L)	0.01	0.01
Electricity (kWh) (sonochemical irradiation)	0.75	0.75
Electricity (kWh) (ethanol distillation)	/	16.65
Output (NPs, g)	5.5	5.5

(c) The sonochemical synthesis process has been modelled for LCIA and the environmental impacts of the synthesis of wCuO and ZnCuO NPs at laboratory scale were evaluated using the CML 2001 impact assessment method and the OpenLCA software.

(d) The interpretation of the results obtained is reported in Sect. 3.1 (see below).

Characterization of NPs and NPs’ suspensions

The characterization of NPs and relative suspensions was performed as reported in [27], with slight modifications. The crystallographic phase of the NPs was confirmed by using X-ray diffraction (XRD) Bruker Inc. (Germany) AXS D8 ADVANCE (voltage 40 kV, monochromatic Cu Kα radiation (λ = 0.15418 nm). Nanopowders were re-suspended in MilliQ (MQ) water to achieve stock suspensions of 10 g/L wCuO or ZnCuO. The stock suspensions were sonicated in a cooling bath for 15 min (SONICA ultrasonic extractor, SOLTEC Srl, Milan, Italy) to obtain a homogeneous distribution of the NPs.

For the NPs’ characterization and toxicological analyses, the test suspensions were prepared by sequentially diluting the stocks in MQ water or embryo solution (ES, whose composition is 100 mg/L NaHCO₃, 100 mg/L Instant Ocean salt, 190 mg/L CaSO₄) to generate nominal concentrations in a range between 0.01and 100 mg/L for wCuO and 0.1–100 mg/L for ZnCuO. Even though these concentrations exceed environmentally relevant levels (our lowest tested concentration is 4–6 orders of magnitude higher than the predicted PEC in water for CuO NMs [28], they were selected to have a sufficiently wide range to determine the LC₅₀ and EC₅₀ values.

For transmission electron microscopy (TEM) analysis, the MQ water suspension (100 mg/L) was sonicated for 1 min, vortexed, and one drop (5 µL) was deposited onto Formvar®-coated 200 mesh copper grids; the excess of water was blotted by filter paper and once air-dried, samples were examined for morphology and size distribution by Jeol-JEM1220 transmission electron microscope.

NPs suspensions (10 and 100 mg/L) prepared in MQ water or ES were used to measure the hydrodynamic diameter (dH) and the polydispersity index (PdI) of wCuO and ZnCuO NPs at 0 and 24 h using Dynamic Light Scattering (DLS) (Malvern Zetasizer, Malvern, UK) analyses, at a scattering angle of θ = 90°. Ζ potentials of both suspensions were also measured.

To quantify the dissolution of wCuO and ZnCuO and estimate the possible contribution of Cu²⁺ and Zn²⁺ to toxicity, NPs suspensions (10 and 100 mg/L) in MQ water were collected immediately (t0) and 24 h (t24) after their preparation. Particle suspensions were then ultrafiltered at 4000 g for 15 min using the centrifuge tubes VIVASPIN20 with a molecular weight cut-off of 10000 Da (Sartorius Stedim Biotech GmbH, Goettingen, Germany). NP-free ultrafiltrates were analysed by Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES) with a Perkin-Elmer Optima 7000DV (Perkin-Elmer, Santa Clara, CA). The analysis was performed in radial viewing mode, and calibration curves were obtained with 0.1, 1.0, 10.0 and 100.0 mg/L standards for Cu and Zn elements (PerkinElmer 1 mg/L Cu and Zn in 2% HNO₃, N9304265). Nitric acid was added to standards and diluted samples (1:10 v/v).

Toxicological analyses

Fish husbandry and breeding

Adult AB wild-type pairs deriving from the European Zebrafish Resource Center (Karlsruhe Institute of Technology, Germany) were bred at the University of Milan-Bicocca zebrafish facility (ethical approval ATS MetroMilano Prot. n. 0020984 – 12/02/2018), in a ZebTec Active Blue circulating system (Tecniplast, Buguggiate, Italy).

Breeders were maintained at 28 °C, pH 7.5, 500 µS and under a 14 and 10 h light–dark cycle. Fish were fed three times per day with Zebrafeed (Sparos Lda, Olhão, Portugal).

The evening before breeding, adult pairs were moved in breeding tanks and separated by sexes during night. In the following morning, the barrier was taken out allowing adults to mate. Fertilized eggs were collected by a strainer within 30 min after mating and rinsed in ES. Embryos at the blastula stage were selected under a stereomicroscope (Zeiss, Germany) between 1- and 3-h post-fertilization (hpf) and used for the Fish Embryo Acute Toxicity (FET) Test.

Fish embryo acute toxicity (FET) test

Acute toxicity of wCuO and ZnCuO NPs and of its ionic form CuSO₄ towards zebrafish embryonic stages was determined as recommended by the OECD Test Guideline n. 236 [21].

For each experimental replicate, 20 embryos were randomly collected from different mates (n = 5), placed in 24-well plates and incubated in 2 mL of ES (control group), CuSO₄ (0.01–1 mg Cu/L) or NP suspensions prepared in ES as above reported (0.01–100 mg/L for wCuO and 0.1–100 mg/L for ZnCuO) in a thermostatic chamber at 26 ± 0.5 °C, under static conditions. In each multiwell plate, the embryos (n = 4) in the last row were considered as the internal control. All treatment and control conditions were performed in biological and technical triplicate.

Control and exposed embryos were observed every 24 h until the end of the test, 96 hpf. Lethal endpoints indicating acute toxicity (coagulation of fertilised eggs, lack of somite formation, lack of detachment of the tail-bud from the yolk sac and lack of heartbeat) were inspected through a stereomicroscope every 24 h.

Besides, additional sublethal endpoints were evaluate during and at the end of the FET test. Malformations, like oedemas and defects to spinal cord, tail, heart and head, were screened and reported at 96 hpf. The hatching rate was checked and registered every 24 h, starting from 48 hpf. The Lethal and Effective Concentrations 50 (LC₅₀ and EC₅₀) were calculated, when possible, starting from the reported lethal and sub-lethal endpoints, respectively.

At the end of the exposure period, a minimum of 10 randomly selected embryos from control and NPs experimental groups of each experimental replicate (n = 30) were anesthetized in 0.016% MS-222 and subsequently fixed in 10% neutral buffered formalin. The head–tail length and the yolk area of each fixed embryos were measured with the digitizing software LasX (Leica Microsystems Srl, Buccinasco, Italy) as indicators of growth retardation, while the head width and the eye diameter were recorded as predictive of sublethal adverse effects related to neurodevelopment.

To elucidate chorion barrier effect, embryotoxicity in terms of lethality and malformation incidence was also evaluated by exposing manually dechorionated embryos from 24 to 96 hpf to wCuO and ZnCuO NPs.

Embryo activity evaluation

Parameters related to bursts of movement (activity) were measured at 24 hpf when embryos are still in the chorion. Spontaneous tail coiling of control and treated embryos (n = 30) was videotaped with the software LasX for 3 min using a camera (Leica DFC450C) coupled to a stereomicroscope (Leica Microsystems Srl, Buccinasco, Italy, M205FA) with brightfield illumination. To minimize the difference in developmental stage between the analysed organisms, given by the developmental progress during recording, different groups of 10 embryos per condition were used at a time. Movies were obtained at a frame rate of 25 frames per second (fps) and exported in MPEG4 format. Raw data were extracted and analyzed by the software DanioScope® (Noldus Inc., Wageningen, The Netherlands). To define the region of interest, a contour was drawn over the chorion. The following parameters were analysed as representative of embryo activity: percentage of the measurement duration in which the embryo was scored as active (“burst activity”), percentage of time of inactivity (“inactivity”), the total duration of all events when the subject was scored as active (“total burst duration”), the total duration of the measurement minus the total burst duration (“inactivity duration”), the number of times the embryo was scored as active (“burst count”), the average duration of all instances when the subject was scored as active (“mean burst duration”) and the number of embryo movements per minute (“burst count per minute”).

Statistical analyses

The mortality percentages after FET tests were calculated as the number of dead embryos scaled to the number of embryos exposed. Similarly, malformation and hatching rates were estimated as the number of malformed or hatched embryos versus the number of embryos survived at the end of FET tests. Data were presented as the average ± Standard Error (SE). Data were tested for homogeneity and normality and one-way analysis of variance (ANOVA) followed by Bonferroni post hoc test was performed.

Embryo cumulative hatching observed during FET tests permitted to calculate through probit method the median Hatching Time (HT₅₀), which corresponds to the hours post fertilization when 50% of the population has hatched, and the 96 hpf-median Effective Concentration (96 hpf-EC₅₀), i.e. the concentration that causes 50% of embryos to fail to hatch at 96 hpf.

All statistical tests were performed with at least 95% confidence using IBM SPSS statistic 26 Software. Results were considered statistically significant if p < 0.05 or p < 0.01.

Results

Life cycle impact assessment of the NPs synthesis process

Eleven midpoint level impact categories have been considered for the environmental impacts estimation: abiotic depletion (ADP), abiotic depletion relative to fossil fuels resources (ADP-f), acidification (AP), eutrophication (EP), freshwater aquatic ecotoxicity (FAETP), global warming (GWP), human toxicity (HTTP), marine aquatic ecotoxicity (MAETP), ozone layer depletion (ODP), photochemical oxidation (POFP) and terrestrial ecotoxicity (TETP). These impact categories were selected because they reflect a comprehensive set of key environmental issues [23, 24].

The relative contribution of each input component to the total impact for every impact category is reported (as %) in Fig. 1.

Fig. 1.

Relative contributions of wCuO (A) and ZnCuO (B) NPs synthesis’ inputs to 11 impact categories

For wCuO nanoparticles (Fig. 1A) electricity’s impact is predominant in 5 impact categories out of 11 (i.e., ADP-f, AP, GWP, ODP and POFP). For ZnCuO NPs’ production (Fig. 1B) the impact of the synthesis is mainly attributable to electricity consumption, as this input majorly contributes to 7 impact categories out of 11 (ADP-f, AP, EP, GWP, MAETP, ODP and POFP) and represents nearly half of the contribution for FAETP and HTTP.

wCuO acetate input represents the second major contribution to 3 impact categories (ADP, HTTP, MAETP) in the synthesis of wCuO NPs (Fig. 1A), while it is reduced to ADP and HTTP in the case of ZnCuO NPs (Fig. 1B). The impact due to deionized water consumption is negligible for both the synthesis processes. Interestingly, ammonium hydroxide consumption (almost) totally contributes to TAETP in both NPs production processes and affects EP and FAETP for nearly or more than 50% in wCuO NPs synthesis, respectively (Fig. 1A). Furthermore, ammonium hydroxide impacts more on the synthesis of wCuO NPs, while it does not have a primary effect in ZnCuO NPs. The additional contribution of zinc acetate input to the ZnCuO synthesis process is negligible.

The absolute impacts associated with the synthesis of 5.5 g of both NPs types are reported in Table 2. Overall, the impacts determined by the ZnCuO NPs synthesis are higher than in the wCuO NPs case, since for all impact categories, except ADP and TAETP, the impact values are one or two orders of magnitude higher. Notably, as regards the impact categories related to toxicity potentials (kg 1,4-DB eq), both the NPs synthesis have a heavy impact on marine aquatic ecotoxicity potential (MAETP), with ZnCuO NPs synthesis having an impact 1 order of magnitude higher that wCuO’s.

Table 2.

Impacts at midpoint level for the synthesis of 5.5 g of wCuO and ZnCuO NPs

Impact category	wCuO NPs	ZnCuO NPs	Unit
ADP	0.0002	0.0002	kg Sb eq
ADP-f	11.5	171.9	MJ
AP	0.006	0.06	kg SO2 eq
EP	0.004	0.018	kg PO4 eq
FAETP	6.5	9.9	kg 1,4-DB eq
GWP	0.9	12.8	kg CO2 eq
HTTP	5.6	9.5	kg 1,4-DB eq
MAETP	3970	20477	kg 1,4-DB eq
ODP	3.6E-08	37.5E-08	kg CFC-11 eq
POFP	0.0002	0.0025	kg C2H4 eq
TETP	1.57	1.45	kg 1,4-DB eq

Characterization of sonochemically produced water-based CuO (wCuO) and Zn-doped CuO (ZnCuO) NPs

Previous works have already investigated the structure of sonochemically prepared MeO NPs [6, 27], thus here are briefly presented the main results obtained for the wCuO and Zn-CuO NPs. The morphology of both MeO NPs was studied by TEM (Fig. 2). The wCuO NPs are organized as leaf-shaped structures in a primary size range between 15 (width) and 60 (length) nm (Fig. 2a). ZnCuO NPs appears as clusters with non-defined morphology consisting of very small NPs in the size range of 2–4 nm [27]. As denoted in Fig. 2b, the XRD pattern of CuO particles clearly demonstrated that all is consistent with the JCPDS data (80–1916) of the CuO monoclinic phase. Figure 2b presents the powder XRD of the material collected after the sonochemical reaction of a mixed solution of copper acetate and zinc acetate in a molar ratio of 3:1 [6]. All the peaks in the XRD spectra are assigned to monoclinic CuO. No peaks related to ZnO or any impurities were observed. It is apparent that the peaks of CuO were slightly shifted and broadened which might indicates on doping of Zn⁺² ions in the unit cell of the monoclinic CuO lattice replacing the Cu⁺² ions. This leads to formation of doped particles, marked as Zn-CuO.

Fig. 2.

Transmission electron microscopy (TEM) images and XRD diffraction patterns of wCuO (a and b, respectively) and ZnCuO (c and d, respectively) NPs

Results from Dynamic Light Scattering (DLS) analyses showed that both the NPs tend to agglomerate in ES, as testified by the extra order of magnitude of the hydrodynamic diameters, with respect to MilliQ water (Table 3). It is noteworthy that this tendency was less marked for the wCuO, which resulted better dispersed at t₀ than Zn doped CuO NPs at 100 mg/L. ZnCuO NPs achieved a better dispersion at t₂₄ in comparison with t₀ even though the NPs clusters remained larger than those of wCuO. The ζ potentials of both NPs were positive and comparable (Table 3), but tended to decrease approaching to zero or even becoming negative in ES, according to the higher ionic strength of this solution (Table 3).

Table 3.

Z-potential and mean hydrodynamic diameter (d_H) of wCuO and ZnCuO NPs

NPs	Primary size (nm)	Concentration (mg/L)	Medium	ζ -potential ± SD (mV)	Time (h)
					0		24
					dH ± SD, nm	Pdi ± SD	dH ± SD, nm	Pdi ± SD
wCuO	25 × 70	100	MQ	+ 31.94 ± 0.93	172.4 ± 17.4	0.193 ± 0.02	145.4 ± 12.5	0.157 ± 0.02
			ES	+ 2.79 ± 0.12	1307.8 ± 393	0.334 ± 0.025	1456.7 ± 184.3	0.223 ± 0.002
		10	MQ	+ 28.98 ± 2.95	184.4 ± 9.9	0.253 ± 0.048	168.4 ± 9.3	0.218 ± 0.005
			ES	−11.73 ± 0.83	n.d	n.d	n.d	n.d
ZnCuO	50 × 75	100	MQ	+ 39.4 ± 1.64	119.23 ± 3.66	0.23 ± 0.02	100.79 ± 5.38	0.18 ± 0.02
			ES	+ 3.66 ± 0.37	3400 ± 475	0.489 ± 0.10	1691.89 ± 217.03	0.34 ± 0.12
		10	MQ	+ 23.26 ± 0.80	121.07 ± 3.67	0.26 ± 0.05	122.47 ± 2.95	0.23 ± 0.03
			ES	n.d	1751 ± 384.42	0.769 ± 0.21	1057.16 ± 177.16	0.64 ± 0.22

d_H = hydrodynamic diameter; SD = standard deviation; MQ = MilliQ water; ES = embryo solution

The results of abiotic dissolution obtained from the ICP-OES analysis demonstrated that both NPs remained stable over time, as the Cu²⁺ ions released were unchanged between 0 and 24 h, indicating that the equilibrium was reached rapidly, namely at time 0 (Table 4). The solubility was proportionally lower at the high NPs concentration (equal to 2.33% of Cu content in 100 mg/L wCuO NPs suspension), compared to the low concentration (equal to 19.8% of Cu content in 10 mg/L), even though the d_H in MilliQ water did not show the formation of larger aggregates as the NP concentration increased (Table 3). Dissolution of Zn from the ZnCuO NPs was below the instrumental detection limit, in line with the limited presence of Zn atoms in the NP. In contrast, the solubilization of Cu from the NP was similar or slightly higher than in wCuO NPs. Analysis with CuSO₄ at the same concentration as Cu in the 100 mg/L wCuO NPs, used as a control, revealed almost complete dissolution of CuSO₄ at t₀ (data not shown).

Table 4.

ICP-OES analyses

NPs	Concentration (mg/L)	Time (h)
		0		24
		Analyte ± SD (mg/L)		Analyte ± SD (mg/L)
		Cu2+	Zn2+	Cu2+	Zn2+
wCuO	100	1.86 ± 0.07	− 2.59	1.80 ± 0.02	− 2.6
wCuO	10	1.58 ± 0.01	− 2.62	1.64 ± 0.03	− 2.62
ZnCuO	100	2.50 ± 0.05	− 1.26 ± 0.02	2.70 ± 0.13	− 1.22 ± 0.03
ZnCuO	10	2.14 ± 0.36	− 2.51	1.89 ± 0.01	− 2.52

SD = Standard Deviation

In vivo acute toxicity on developing zebrafish (Danio rerio)

FET results

No lethal effects were observed at any of the tested concentrations of wCuO NPs, while embryos exposed to the highest concentration (100 mg/L) of ZnCuO NPs showed a statistically significant mortality rate (> 35%) compared to the control group (Fig. 3). This difference is further supported by the extrapolated 96 hpf-LC₅₀ values, with ZnCuO NPs having a lower and calculable LC₅₀ value (equal to 123 mg/L) than wCuO NPs (which was >  > 100 mg/L, i.e., the highest concentration tested). No significant differences were observed in the malformation rates of the two treatments (Fig. 3), thus, it was not possible to estimate the EC₅₀ based on phenotypic abnormalities. No significant increase in lethality and malformation incidence was observed even in dechorionated embryos exposed from 24 to 96 hpf to the same concentrations of wCuO and ZnCuO NPs tested in conventional FET (Fig. S1, supplementary information). In contrast, ionic form of copper was highly embryotoxic to zebrafish, with a 96 hpf-LC₅₀ value of 0.206 (0.167–0.270) mg Cu/L (Fig. S2, supplementary information).

Fig. 3.

Comparative embryotoxicity of wCuO (a) and ZnCuO NPs (b) in 96 h post fertilization (hpf) zebrafish embryos expressed as mortality and malformation rates (%). All values are given as mean ± SE of three independent assays (embryos n = 60). *p < 0.05 vs control (One-way ANOVA + Bonferroni method). Median Lethal Concentration (LC₅₀) at 96 hpf is reported for each NP

A significant (p < 0.01) suppression of hatching at 72 hpf was observed in embryos treated with wCuO NPs, even at the lowest comparison concentration tested (0.1 mg/L), with no recovery of the effect at the end of the test (96 hpf) (Figs. 4 and 5). By adding two lower concentrations of wCuO NPs (0.01 and 0.05 mg/L), it was possible to determine the wCuO median Hatching Concentration (HC₅₀) value equal to 0.026 mg/L. Indeed, embryos treated with wCuO NPs at 0.01 mg/L showed a delayed hatching process but were fully chorion free by 96 hpf (Fig. 4a). An impact on hatching was also noted in ZnCuO-treated embryos, although to a lesser extent, and a partial recovery at 96 hpf was seen at the lowest and intermediate concentrations tested (0.1, 1, 10 mg/L), consistent with the higher 96 hpf-HC₅₀ calculated for ZnCuO NPs (0.119 mg/L) in comparison with that of wCuO NPs (Fig. 4b). The hatching was also strongly inhibited in copper sulphate exposed embryos, the cumulative hatching rate at 96 hpf being approximately 28% at 0.01 mg Cu/L (Fig. S3, Supplementary Information).

Fig. 4.

Hatching rates of zebrafish embryos following exposure at increasing concentrations of wCuO (a) and ZnCuO (b) NPs. All results are shown as mean ± SE of three independent assays (embryos n = 60). *p < 0.05 vs control, **p < 0.01 vs control, ^##p < 0.01 versus 0.01 mg/L (One-way ANOVA + Bonferroni method). Median Hatching Concentration (HC₅₀) at 96 hpf is reported for each NP

Fig. 5.

Representative images of control zebrafish larvae (96 hpf) (a) and those exposed to increasing concentrations (0.1, 1, 10, and 100 mg/L) of wCuO (b-e) and ZnCuO (f-i) NPs for a total of 96 h after collection of fertilized eggs. Larvae were taken from groups of three independent assays (n = 60). Scale bars: 500 µm

The estimated average times needed for 50% of the embryos to hatch (median Hatching Times 50, HT₅₀) related to 0.01 and 0.05 mg/L for wCuO and 0.1–100 mg/L for ZnCuO NPs were significantly higher and showed a concentration-dependent hatching delay when compared to controls (Table 5). Notably, wCuO HT₅₀ values fall beyond the overall duration of the FET test (96 hpf) already at the 0.05 mg/L treatment (HT₅₀: 177 hpf) and after this concentration the values were too high to be considered realistic (thus noted as n.d., Table 5).

Table 5.

Median hatching time (HT₅₀) in zebrafish embryos exposed to wCuO and ZnCuO NPs

	HT50 (hpf)
Treatment (mg/L)	wCuO	ZnCuO
Control	78 (75—80)
0.01	83 (80—86)	n.d
0.05	177 (122—>  > 177)*	n.d
0.1	n.d	104 (98—112)*
1	n.d	99 (95—104)*
10	n.d	102 (98—107)*
100	n.d	125 (110—211)*

n.d. denotes not determined, HT₅₀ stands for time, expressed as hours post fertilization (hpf), required for 50% hatching of embryos. *p < 0.05 vs control group

Larvae from control and treatment groups collected and fixed in 10% neutral buffered formalin after the end of the FET test (96 hpf) were analysed for morphometric parameters.

Many of the wCuO and ZnCuO-exposed groups significantly differed from controls for the measurements considered in this study (Fig. 6). Overall, a decrease in the mean value of the parameters was observed for wCuO NPs compared to the control larvae, especially at higher concentrations. Specifically, for yolk area and head width (Fig. 6b and c, respectively), wCuO concentrations of 10 and 100 mg/L induced significant reductions with respect to the control, but also towards 0.1 and 1 mg/L treatments. The decrease in embryo length, head width and eye diameter compared to control was also registered in the ZnCuO-exposed groups, although the concentration-dependence was not as evident as in CuO treated embryos (Fig. 6a, c and d). Instead, a trend towards an increase in yolk area was noted, especially at the highest concentration of ZnCuO NPs (Fig. 6b).

Fig. 6.

Morphometric measurements of 96 hpf zebrafish larvae exposed to increasing concentrations of wCuO and ZnCuO NPs, following Fish Embryo acute Toxicity test exposure. All values are given as mean ± SE of three independent assays (n = 30). * p < 0.05 vs control, **p < 0.01 vs control, ^##p < 0.01 versus 0.1 mg/L, ^§§p < 0.01 versus 1 mg/L, (One-way ANOVA + Bonferroni method)

Evaluation of embryonic activity at early developmental stages

In order to detect any signs of developmental neurotoxicity and determine if the inhibition of the hatching process was a result of a potential change in the activity of pre-hatching embryos, we examined the motility-related factors in the chorion of individuals exposed to wCuO and ZnCuO NPs. We focused on the lowest concentrations at which the effect was observed (0.1 and 1 mg/L) for both types of NPs. Video recordings and analyses were performed at 24 hpf.

Compared to control values, none of the analysed parameters, except for burst count per minute for 1 mg/L wCuO, showed any significant differences for the tested concentrations of both NPs (Table 6). However, a general trend towards a reduced total coiling activity and duration was observed for wCuO NPs. Instead, it was registered an increase in the coiling frequency (burst count per minute) which turned out significant (p < 0.05) in 1 mg/L wCuO NPs’ embryos treated (Table 6).

Table 6.

Activity parameters of pre-hatching zebrafish embryos (24 hpf) exposed to 0.1, 1 mg/L of wCuO and ZnCuO NPs

NPs	Treatment group	Burst activity (%)	Inactivity (%)	Total burst duration (s)	Mean burst duration (s)	Inactivity duration (s)	Burst count	Burst count/min
wCuO	Control	64 ± 3	35 ± 3	8.28 ± 0.37	0.76 ± 0.10	4.64 ± 0.37	13 ± 1	4.3 ± 0.21
	0.1 mg/L	62 ± 3	37 ± 3	8.05 ± 0.37	0.66 ± 0.07	4.87 ± 0.37	14 ± 1	4.66 ± 0.21
	1 mg/L	59 ± 3	41 ± 3	7.11 ± 0.48	0.52 ± 0.05	5.03 ± 0.45	16 ± 1	5.3 ± 0.21*
ZnCuO	Control	44 ± 3	56 ± 3	5.71 ± 0.39	0.43 ± 0.03	7.21 ± 0.39	14 ± 1	4.66 ± 0.21
	0.1 mg/L	39 ± 4	61 ± 4	5.09 ± 0.48	0.48 ± 0.11	7.83 ± 0.48	13 ± 1	4.16 ± 0.29
	1 mg/L	45 ± 3	55 ± 3	5.81 ± 0.44	0.39 ± 0.04	7.11 ± 0.44	16 ± 1	5.38 ± 0.29

All values are given as mean ± SE (n = 30 individuals for each treatment group). *p < 0.05 vs control (One-way ANOVA + Bonferroni method)

Discussion

In the framework of EU, projects aimed at developing nano-enabled strategies to produce textiles, water filters and other contact materials with enhanced antimicrobial properties, coating technologies based on the use of CuO NPs are being developed. In this perspective, to contribute to the safe and sustainable development of new advanced antimicrobial materials, the hazardous properties and the environmental impacts of two CuO nanoforms (wCuO and ZnCuO NPs), sonochemically synthesized to be used as antimicrobial coating agents, were tested using a zebrafish embryo toxicity test and LCA methodology respectively. With this approach, our work aimed at elucidating whether and how the synthesis conditions and metal doping of these novel non-commercial antimicrobial CuO NPs could affect the NPs properties and their toxicological profile, as well as the effects on the different environmental impact categories represented in standard LCA. The application of the SSbD approach in this work is represented by the investigation on the safety and environmental sustainability issues underlying the viability of designing antimicrobial NPs for products enabling. The methodology and results outlined offer a robust basis for the SSbD framework application, providing a pathway for the development of NEPs that balance both the aspects. This study serves as a model for integrating multidisciplinary assessments to inform the development of safe and sustainable technologies.

From the FET test results, it emerged that neither wCuO nor Zn-doped CuO NPs showed any severe developmental hazardous properties, since no acute embryotoxic or teratogenic effects were evidenced, except for ZnCuO NPs at the highest concentration. At 100 mg/L, in fact, ZnCuO NPs induced a significant mortality in zebrafish embryos, but this value is orders of magnitude higher than the expected environmental concentrations for Cu-based NMs, which are predicted to be in the range of ng/L [28]. About the effects of doping metal oxide NPs with other metals there is still conflicting information and results. On one hand, it has been acknowledged that the hazard of a NP can be reduced by surface modification (capping or coating) and doping, as these processes can interfere with the release of metal ions responsible for the generation of reactive oxygen species (ROS) [14, 29]. On the other hand, studies have indicated that metal doping of NPs increases their structural defects, thereby enhancing their ability to induce reactive oxygen species (ROS) production, which improve their effectiveness as antibacterial agents [6, 15] but at the same time could worsen its toxicity toward non-target organisms. In fact, ROS are known to increase oxidative stress and eventually lead to oxidative damage in cells and tissues [30]. The production of ROS species by the sonochemically prepared MeO of this work was already studied [6, 27]. ZnCuO NPs are capable of producing a larger amount of ROS as compared to the pure CuO or ZnO NPs. The results point out to the higher antimicrobial activity of the doped materials, as compared to ZnO and CuO NPs. In pure CuO, the amount of OH• is lower than 15%, while ZnO produces only OH radicals and no superoxide anions. OH radicals, in addition to the singlet oxygen, may be formed through the interaction of Zn²⁺ and O₂⁻ in the Zn–CuO unit cell. The results point out to the higher antimicrobial activity of the doped materials, as compared to ZnO and CuO NPs. The higher activity is due to the higher total ROS production and subsequently higher OH radical, superoxide anions, and singlet oxygen formation by Zn–CuO NPs.

However, it can be assumed an overall good safety profile for the two CuO-based NPs, at least when considering the acute lethal and sublethal effects in terms of abnormal development, since LC₅₀ values can be estimated to be > 100 mg/L and EC₅₀ based on the morphological observations were not calculable for both NPs.

Similar results were previously obtained also in amphibian embryos exposed to ZnO, CuO and ZnCuO sonochemically synthesized in ethanol solution [27], confirming that these materials do not pose a significant harm to vertebrate developing embryos. Our result is significant because, despite the wide range of variations, the existing literature on the developmental effects of CuO NPs and their functional modifications on Danio rerio indicates an impact on survival, with LC₅₀ values below 100 mg/L and the occurrence of abnormal phenotypes in embryos exposed to commercial CuO NPs [31–35]. These abnormalities include axial (spinal cord malformation, end tail malformation, scoliosis, rachischisis and tail malformation) and non-axial (head malformation, smaller eyes, heart malformation, yolk deformity and growth retardation) malformations. Significative lower survival rates and dose-dependent abnormal phenotypes have been found in zebrafish embryos exposed to commercial CuO NPs [31, 33, 34], already starting from 12.5 mg/L [31]. Ganesan and collaborators [33] found a LC₅₀ of 64 mg/L for commercial CuO NPs, while the study of Xu et al. [34] observed a mortality of 60% in embryos exposed to 50 mg/L of non-commercial CuO nano-powders already at 24 hpf, which increased up to approximately 80% at the end of the treatment (120 hpf).

On different aquatic species, a high variability in CuO NPs toxicity was also noted, depending on the specific sensitivity toward NPs and the dissolved Cu ions and to their uptake capacity [32]. These authors reported that Cu NP exposure is able to inhibit algal growth, D. magna survival, and zebrafish hatching, more efficiently than the equivalent concentrations of dissolved Cu and even of CuO NPs. From our results, it emerged that the embryo hatching was the main target of wCuO and ZnCuO NPs, with the former being the most effective although to a much lower extent than dissolved Cu²⁺ (see SI). Copper ions and CuO NPs have been shown to significantly suppress hatching of zebrafish in a dose-dependent pattern by affecting embryonic motility as a consequence of oxidative stress [36]. On the contrary, [37, 38] mechanistically linked the hatching interference of CuO to an inhibition of the enzyme responsible for hatching (zebrafish hatching enzyme, ZHE1), possibly involving ligation of critical histidines (metal-sensitive binding sites) in ZHE1 active centre by the dissolved metal ions. Our results showed that Zn-doping reduced the NPs impact on hatching. A potential explanation for the observed result may involve the interference caused by zinc metal doping on the release and availability of copper ions from CuO NPs, which could lead to a reduced inhibition of the hatching enzyme. This hypothesis aligns with the results reported by [14], who documented a progressive decrease in hatching interference in zebrafish embryos with increasing levels of Fe doping in CuO NPs, attributing this effect to a lower release of Cu²⁺. However, our ICP results do not indicate a reduced release of Cu²⁺ from zinc-doped NPs, prompting us to discount this hypothesis. Additionally, in a saline solution such as the embryo medium, it is likely that these ions are chelated by the presence of chloride ions or other negatively charged ions, similar to what occurs in natural water. Furthermore, the presence of salts contributes to the agglomeration of NPs by interacting with their surface charges, thereby altering their stability.

However, the interaction of various MeO NPs with the chemistry of microenvironments of biological matrices, such as chorion, perivitelline fluid, and epithelium, can influence their accumulation and the release of soluble ions at the toxicity site, thereby affecting hatching to varying degrees [39, 40]. The role of the chorion in modulating the toxicity of MeO NPs is clearly demonstrated in embryos with an intact chorion exposed to ZnCuO NPs, compared to dechorionated embryos, where no mortality was observed even at the highest tested concentration.

Considering the embryo motility within the chorion, our results did not reveal a significant impact of CuO and ZnCuO NPs on, excluding the coil tail activity as a factor responsible for the embryos’ inability to breach through their protective shell. The surface of chorion membrane was visibly covered by a substantial accumulation of CuO and ZnCuO NPs agglomerates, suggesting that the chorion may reduce NP–embryo interaction during the early stages of development by adsorbing these particles and initially acting as a physical barrier to their entry [35, 41]. These deposits could lead to the hardening of the chorion and the filling of the pore channels, ultimately decreasing the passage of oxygen and thereby causing delays in development and hatching, similar to the effects observed also with non-metallic particles such as microplastics [42, 43]. The data from the morphometric analysis support a general delay in embryonic growth, which was particularly pronounced for CuO NPs at higher concentrations; however, the specific mechanisms through which these NPs significantly affect the hatching process were beyond the scope of the current study and remain to be elucidated.

In conclusion, considering the above evidences, the hatching rate and timing can be considered the most predictive endpoints for the CuO-mediated toxicity in zebrafish embryos, and the NPs doping with Zn seemed to be able to alleviate this adverse effect. This occurs independently of the solubility of the NPs and thus should be reasonably attributable to the modulation of the shape and surface reactivity of the doped NPs.

The environmental impact assessment of the production processes for antimicrobial ZnCuO and wCuO NPs (technical unit 5.5 g) overall reveals significant differences in their contributions to the considered midpoint impact categories. The results highlight the dominant role of electricity consumption and specific chemical inputs in shaping the environmental footprint of these nanoparticles’ synthesis.

Electricity usage emerged as the primary contributor to environmental impacts in both ZnCuO and wCuO nanoparticles production. Electricity is commonly found to be one of the most impacting inputs in the synthesis process of metallic nanoparticles, regardless of the method used. This is also confirmed by other studies, such as [44], who analysed the environmental impact of producing metallic nanoparticles using the arc/spark method and found that for most of the Cu and Zn NPs synthesis-setups the largest contribution to the overall environmental impact was in fact given by electricity consumption. Also [45] showed that physical and, interestingly, biological synthesis routes for iron-oxide NPs exhibited high environmental impacts due to their higher energy and material requirements, when compared to chemical methods. These evidences indicate that the energy-intensive nature of different synthesis processes to obtain metal and MeO NPs is a critical factor in the overall environmental burden. The reliance on electricity suggests that strategies aimed at reducing energy consumption, such as optimizing synthesis conditions, shifting the attention to natural reagents or integrating renewable energy sources, could significantly mitigate these impacts [46]. Indeed, nanotechnologies have been included in the group of innovative materials and strategies that could help in the sustainable and circular growth [47]. Copper acetate was the second most significant contributor after electricity in the production of wCuO NPs, particularly in the ADP, HTTP and MAETP categories, due to copper extraction from abiotic resources and the release of pollutants (e.g., PM₁₀, PM_2.5, CO₂, SO₂, NO_x and metals) along the related procedures [48]. For ZnCuO NPs, however, the impact given by this chemical was reduced to ADP and HTTP, with zinc acetate showing a negligible additional impact. Considered together, the results described so far and the fact that the overall impacts for ZnCuO NPs are higher when compared to wCuO NPs across most categories, suggest that the discriminant to the alteration of the environmental burden in the synthesis of ZnCuO NPs is definitely represented by the electricity that is being used for the ethanol distillation, rather than by the inclusion of zinc as an input. Electricity generation requires fossil fuels resources, which explains the contribution to ADP-f category. The contribution to toxicity-related categories can be linked to the release of wastes and toxic compounds (e.g., heavy metals and polycyclic aromatic compounds) into the water resources following electricity generation and ethanol production [49]. Furthermore, electricity exploitation involves, as direct and indirect consequences, emission of greenhouse gases, particulate matter, SOx compounds, flora and fauna alteration (to name a few) and this is why the categories GWP, ODP, AP and POFP are so heavily affected [50].

Ammonium hydroxide was identified as a major contributor to TETP in both nanoparticles synthesis processes, as well as having substantial effects on EP and FAETP in the wCuO NPs synthesis. Ammonium hydroxide (NH₄OH) is the aqueous solution of ammonia, which, when in water, is in fact a combination of NH₃, NH₄⁺, and OH⁻. Ammonium hydroxide can harm both water and soil quality if not properly disposed. It can cause eutrophication in aqueous environments, leading to excessive algae growth and oxygen depletion, which eventually harms aquatic organisms. Furthermore, high concentrations in soil can alter its chemical composition, be toxic to plants and disrupt soil ecosystems [51, 52]. The influence of ammonium hydroxide on these categories highlights the need for careful management and potential substitution of this input to reduce the associated environmental impacts.

Within the framework of circular economy, several potential optimization strategies can be proposed to reduce environmental impacts during the synthesis of the two analysed NPs, and could be considered for scaling up sustainable NPs production. The use of alternative energy sources in our investigated sonochemical processes, such as renewable energy, may play a significant role in specific environmental impact reduction. Implementing methods to recover and reuse synthesis precursors, such as copper or zinc acetate, can contribute in lowering raw material extraction, thus reducing associated environmental impacts. Also, exploring alternative less impactful precursors could lead to a decreased burden. Moreover, further studies on process upscale may reveal higher process efficiencies which are expected to provide reduced direct energy requirement, and thus reduced environmental impacts [53].

The proposed LCA study has limitations due to the exclusion of the use and end-of-life stages, potentially leading to oversimplified analyses and misleading conclusions about product sustainability. To improve LCA, transparent boundary reporting, inclusion of more lifecycle stages, sensitivity analyses, and stakeholders’ engagement should be used [54–56]. Additionally, current LCA datasets lack nano-specific effect factors, limiting the assessment of NPs emissions. For instance, CuO NPs emissions cannot be modelled accurately due to a lack of nano-specific data, as their environmental effects differ significantly from bulk CuO. Research is needed with this regard, aimed at addressing the gaps with the determination of nano-specific effect factors, especially for toxicity and eco-toxicity, is essential for harmonizing LCA methodologies and advancing safety and sustainability assessments of NMs [57].

For future works, expanding the system boundaries and redefinition of the Functional Unit would allow to obtain a better comparative basis to assess environmental impacts associated to different process recipes, processing routes and NPs species. Indeed, considering the complete life cycle and the antimicrobial functionality level obtained for different synthesised NPs species would allow a more suitable allocation of impacts associated to the use of chemicals and to energy consumption. This would help in revealing the antimicrobial agent’s best sustainable synthesis route per unit functionality attainable.

Conclusions

Our study suggests that ZnCuO NPs are a valid antimicrobial alternative in terms of safety, though Life Cycle Assessment findings indicate that their production has greater environmental impacts compared to wCuO, which appears more sustainable for antibacterial applications. These controversial results have significant implications for advancing NMs development and regulation, supporting the need for further enhancing integrated methodologies to assess both safety and sustainability at an early stage of new materials design. Indeed, the insights resulting from such approach can guide regulators in creating clearer NM development guidelines. In this sense are oriented the EU strategies for safe and sustainable new chemicals and materials and the current projects aimed at implementing the SSbD framework. These efforts, especially in the field of nanotechnologies, will contribute in introducing key actionable recommendations to reduce the safety and environmental burdens of NPs production, such as energy use optimization, alternative energy sources exploitation and new recipes exploration by retaining the same antimicrobial functionality performance. These actions may be a further improvement route to target and reduce the overall process impacts, particularly in terms of toxicity potentials and non-renewable resource depletion. This study provides valuable insights into the development of safer and more sustainable antimicrobial NMs, and underscores the importance of balancing safety with environmental considerations, aligning production processes with the SSbD framework to optimize outcomes. These results provide a crucial foundation for the advancement of nano-enabled technologies with both applicative and environmental benefits.

Finally, while this study provided insights on hazards and environmental impacts at lab-scale level, future investigations on processes upscaling will help to extrapolate the results at industrial level, in order to achieve a better assessment of the product safety and sustainability profile for its future industrial and commercial exploitation.

Author contributions

B.N.—toxicity assessment and LCA expertise, conceptualization, investigation, methodology, formal analyses, data curation, writing (original draft preparation, review and editing). P.F.—toxicity assessment expertise, conceptualization, investigation, methodology, formal analyses, data curation, writing (original draft preparation, review and editing). C.D.—toxicity assessment expertise, investigation, methodology, data curation, formal analyses, writing (review and editing). A.A., M.C.—LCA expertise, investigation, methodology, data curation, formal analyses, writing (review and editing). M.S.—Transmission electron microscopy expertise, methodology, formal analyses, data curation. I.P.—NP synthesis and physicochemical characterization expertise, methodology, formal analyses, data curation, writing (original draft preparation, review and editing). P.B., A.C.—toxicity assessment expertise, conceptualization, investigation, methodology, formal analyses, data curation, writing (original draft preparation, review and editing), supervision. M.P.—LCA expertise, project administration, funding acquisition, writing (review and editing). P.M.—toxicity assessment expertise, conceptualization, project administration, funding acquisition, writing (review and editing). All authors have read and agreed to the published version of the manuscript.

Funding

The authors would like to thank the EU and Italian MUR for funding, in the frame of the collaborative international consortium AMROCE financed under the ERA-NET AquaticPollutants Joint Transnational Call (GA n.869178). This ERA-NET is an integral part of the activities developed by the Water, Oceans and AMR Joint Programming Initiatives. This work has received funding from the European Union’s Horizon Europe research and innovation program under grant agreement N.101138414—INTEGRANO project.

Data availability

No/Not applicable (this manuscript does not report data generation or analysis). The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information files. Should any raw data files be needed in another format they are available from the corresponding author upon reasonable request.

Declarations

Ethics statement

The experiments were performed on embryos between 0 and 96 hpf, thus considered as an in vitro model according to the Italian animal welfare regulation (L.D. 4 March 2014, n. 26. Implementation of Directive 2010/63/EU on the protection of animals used for scientific purposes).

Competing interests

The authors declare no competing interests.