Controllable Cu⁺/Cu²⁺ ratio for the gas-sensing property of (Na, Cu) co-doped ZnO investigated by EPR and SPV

Abstract

Herein, (Na, Cu) co-doped ZnO nanocrystals were prepared using the sol–gel method. By carefully balancing the Cu⁺/Cu²⁺ ratio through Na doping in Zn_0.95Cu_0.05O-based sensors, the gas-sensing activity can be significantly enhanced. Interestingly, we found that increasing Na doping results in the transformation of Cu²⁺ to Cu⁺ in the copper ion valence state. Furthermore, the XPS results indicate that Na⁺ ions increase the oxygen vacancies of the samples, which is in agreement with the electron paramagnetic resonance (EPR) results. The surface photovoltage (SPV) spectra indicate that (Na, Cu) co-doped ZnO nanocrystals with a high Cu⁺/Cu²⁺ ratio exhibit a high positive SPV response, demonstrating the excellent separation efficiency of photogenerated charges. Oxygen vacancies and the transformation of Cu²⁺ to Cu⁺ are presumed to be the driving factors responsible for UV light-activated NO₂ gas-sensing improvement in this study. This work demonstrates a strategy for improving the gas-sensing efficiency of ZnO-based sensors through leveraging the variable valence states of incorporated metal species.

---

1. Introduction

Zinc oxide (ZnO) is a significant n-type semiconductor material with a wide band gap of 3.37 eV,^1,2 which is widely used in the development of diode devices,^3–5 photodetectors,⁶ and antibacterial agents.^7,8 Inorganic metal oxides like TiO₂,^9,10 MgO,¹¹ ZnO,^12–14 and CuO¹⁵ exhibit biocompatibility as well as excellent chemical and thermal stability. Doping ZnO with transition metals such as Mg, Co, Cu, and Ni has been shown to enhance its physical, chemical, and antibacterial properties, thereby expanding its potential for various applications.^16,17 Using Cu-doped polycrystalline ZnO deposited on glass substrates, it has been found that refractive indices increase and forbidden energy gap values decrease with increasing Cu dopant concentration. The enhanced degradation performance of Cu-doped ZnO synthesized with a facile solution route can be attributed to hierarchical nanostructures and the formation of an acceptor level (Cu²⁺–Cu⁺), compared with pristine ZnO.¹⁹ According to the abovementioned literature, the valence state of Cu has a strong influence on the microstructure performance of the samples. Brahma²⁰ synthesized Cu-doped ZnO nanowires via chemical vapor deposition and demonstrated striking responses to acetone gas at 1 ppm at room temperature for only the p-type Cu-doped ZnO. Interestingly, the variation of n-type Cu-doped ZnO to p-type is due to the successful substitution of Cu⁺ for Zn²⁺ lattice sites via Cu doping. Ganesh systematically studied the Cu transition metal for enhancing the response and recovery time of the gas sensor, and Cu-doped ZnO (6 wt%) nanoflowers showed enhanced selectivity toward ammonia at a 10-ppm concentration.²¹ One unresolved issue is that the effects of tuning and probing Cu valence states remain insufficiently explored. Fabrication of Na-doped ZnO nanowires, as potential materials of blue emission, via a thermal decomposition route led to a strong photoluminescence band with a major peak at 420 nm.²² Na as a substitutional group-I element is a shallow acceptor for the Na-doped ZnO, which is able to inhibit enzymes and facilitate the generation of reactive oxygen species. In Sáaedi's study,²³ undoped and Na-doped ZnO nanorods were prepared, and their gas-sensing performance against ethanol was evaluated. Compared with pure ZnO, surface oxygen vacancies induced by Na doping are presumed to be a driving factor behind the improvement of gas-sensing efficiency in Na-doped ZnO nanorods. Meanwhile, Na-doped ZnO prepared using a simple precipitation method exhibited a highly efficient photocatalytic performance for the methylene blue dye compared with pure ZnO and is considered a promising photocatalyst based on density functional theory calculations.²⁴ Jaisutti²⁵ argued that p-type Na-doped ZnO nanoflowers with hierarchical nanosheets were synthesized as acetone gas sensors with a fast response time of 18 s (100 ppm), which enables the diagnosis of various diseases, including diabetes, from exhaled breath. In this regard, Na-doped ZnO is worthy of further study to illuminate the behavior of sodium in ZnO.

To date, many studies on Cu- or Na-doped ZnO have focused on the influence of Cu or Na doping alone on photocatalytic activities or on the nanostructure and electrical properties at different doping concentrations. As far as we know, there are virtually no reports on the interfacial effects of Na ion doping on the gas-sensing performance of the Zn_0.95−xCu_0.05Na_xO nanocrystalline. In our study, we propose whether it is possible to further enhance gas-sensing performance by tuning the Cu⁺/Cu²⁺ ratio as a facile way to enrich the surface active sites of (Na, Cu) co-doped ZnO. Specifically, by precisely controlling the doping concentration of Na⁺ ions to modulate the electron transfer between Cu⁺ and Cu²⁺, and by leveraging the synergistic redox effects arising from heterostructure formation and surface deep-level defects, the photo-excited carrier lifetime is significantly prolonged, leading to enhanced NO₂ gas-sensing performance.¹⁸

2. Experimental

2.1 Synthesis of Zn_0.95−xCu_0.05Na_x nanocrystals

Na-doped Zn_0.95Cu_0.05O nanocrystals were prepared using the sol–gel method. Sodium acetate trihydrate, zinc nitrate hexahydrate, and copper acetate monohydrate were the individual sources of Na, Zn and Cu, respectively. The above chemicals were added to a suitable alcohol in a beaker to form a homogeneous precursor solution, which was incubated in a water bath at 60 °C for 2 hours under magnetic stirring. A stabilizer (2-ml ethanolamine) was added to the precursor, and the pH of the precursor was adjusted to 8 using glacial acetic acid or ammonia. The solution was incubated for 72 hours under dark conditions at room temperature. The resultant gel was dried at 80 °C for 8 hours in a dryness box and then calcined in a muffle furnace at 800 °C for 6 hours in an air atmosphere. The fabricated samples were denoted as Zn_0.95−xCu_0.05Na_xO nanocrystals (x = 0, 0.005, 0.01, 0.02, 0.05, and 0.1), corresponding to the molar ratios of CH₃COONa to Zn(NO₃)₂.

2.2 Sample characterization

An X-ray automatic diffractometer (DX-2000) was used to characterize the nanocrystals' phase compositions with a step size of 0.02°. The surface morphology of the samples was investigated using a transmission electron microscope (JEOL JEM2100PLUS) operating at an acceleration voltage of 200 kV and a scanning electron microscope (HITACHI Regilus8100). The surface chemical composition was identified using an X-ray photoelectron spectrometer (VG ESCALAB 210) calibrated to the C 1s peak of 285.0 eV for the binding energies. The spectrum decomposition was analyzed using the Gaussian functions of the XPS PEAK program. N₂ adsorption and desorption isotherms were measured using the Micromeritics ASAP2460 Surface Area and Porosimetry Analyzer, where all polymer samples were degassed at 110 °C for 10 h under a vacuum of 10⁻⁵ bar prior to analysis. The BET surface area was evaluated based on the N₂ adsorption and desorption isotherms using the BET equation. Electron paramagnetic resonance (EPR) measurements were made using a Germany E500 spectrometer to further characterize the intrinsic defects of Zn_0.95−xCu_0.05Na_xO nanocrystals at room temperature. The surface photovoltage (SPV) spectra were measured using a custom-made instrument containing a lock-in amplifier (SR830-DSP), a light chopper (SR540) and a photovoltaic cell in our laboratory.

2.3 Gas-sensing measurements

Au interdigitated electrodes on an Si substrate were fabricated using a micro-electro-mechanical system (MEMS) technology.²⁶ Fig. 1(a) presents the structure and cross-section of the as-prepared sensors. An interdigitated electrode was fabricated through the following steps: first, a 500-nm-thick silicon dioxide was thermally grown on Si (100) as AN insulating layer. Second, a 200-nm-thick titanium and a 500-nm-thick gold were deposited via thermal evaporation. Third, the IDE with a finger width of 50 µm and a gap width of 50 µm was prepared via conventional photolithography and the lift-off method. For cleaning the Si substrate, acetone, ethanol, and deionized water were added in turn, and nitrogen gas stream was used to dry the substrate. Ethanol suspensions containing fabricated nanocrystals (3.0 mg ml⁻¹) were added to the Si substrate via spin coating. UV LED light sources (365 nm, Taiwan Light Macro Chip) and the Keithley 2700 data acquisition system were used for gas-sensing measurement. The sensor response was calculated as (R_g – R₀)/R₀, where R_g is the real-time resistance exposure to NO₂ and R₀ is the real-time resistance of dry air from a compressed cylinder.²⁷ Sensitivity was defined as the slope of the response-concentration fitting curve.²⁸

Fig. 1 (a) Structure and cross-section of the as-prepared sensors; (b) XRD pattern of the Zn_0.95−xCu_0.05Na_xO nanocrystals (x = 0, 0.005, 0.01, 0.02, 0.05, and 0.1).

3. Results and discussion

3.1 Material structural studies

To uncover the physical basis for the gas sensing and antibacterial properties of the nanocrystals, the microstructure of Zn_0.95−xCu_0.05Na_xO nanocrystals was characterized via X-ray diffraction (XRD). Two sets of diffraction peaks are shown in Fig. 1(b). The strong and sharp diffraction peaks of hexagonal wurtzite ZnO are visible in the XRD pattern, providing evidence that the fabricated samples are well-crystallized. Previously, we reported (Co, Cu)-co-doped ZnO and did not observe CuO peak,²⁹ which could be due to the high-temperature treatment and long treatment time. This work presents a strategy for synthesizing oxide composite nanomaterials. Meanwhile, there were slight shifts in the peak positions of the Zn_0.95−xCu_0.05Na_xO nanocrystals with the introduction of Na. The Na ions prefer to aggregate at interstitial sites because the ionic radius of Na⁺ (1.02 Å) is larger than that of Zn²⁺ (0.74 Å).³⁰ Cu²⁺ (0.73 Å) tends to occupy Zn ion sites because of the similar radii of Cu and Zn. Therefore, these data confirm that Na and Cu are incorporated into the Zn–O lattice.

The surface morphology of Zn_0.95−xCu_0.05Na_xO nanocrystals was characterized using scanning electron microscopy (SEM) images depicted in Fig. 2(a)–(f). The sample morphologies were not appreciably changed by doping with Na and Cu, which appeared quasi spherical. Moderate Na doping seemed to have improved crystal nanoparticle uniformity. In the HRTEM images of Zn_0.9Cu_0.05Na_0.05O in Fig. 2(g) and (h), the nanomaterial exhibits large clusters of 100-500-nm particles due to the high treatment temperature, in which the interplanar lattice spacings of 0.235 and 0.283 nm belong to the (111) and (100) planes of CuO and ZnO, respectively. These findings are in agreement with the XPS results. XPS is a sensitive technique for assessing the electronic structure of the surfaces of solids.³¹ In the high-resolution Zn2p XPS spectra for Zn_0.95−xCu_0.05Na_xO nanocrystals with different Na concentrations, no significant variation in Zn chemical states from the typical Zn²⁺ state was observed for any sample (Fig. 3). Meanwhile, the Na 1s peaks of the 1072.1 eV binding energy of the XPS spectrum of Zn_0.95−xCu_0.05Na_xO did not change, regardless of whether the Na concentration was variable (Fig. 4). Remarkably, we observed that the O 1s XPS spectrum profile was quite asymmetric (Fig. 5). The spectrum could be resolved into two components centered at ∼530.1 eV and ∼531.9 eV using Gaussian functions. The low binding energy component of the O 1s spectrum is attributed to lattice oxygen in the ZnO matrix, while the higher binding energy peak corresponds to surface hydroxyl groups (OH–) associated with oxygen vacancies.³² It is worth noting that the high binding energy peak is in accordance with the Zn–O bond increasing in intensity with Na doping, suggesting that higher Na concentrations favor O²⁻ ion formation in the wurtzite structure of the hexagonal Zn²⁺ ion array. The Cu 2p_3/2 spectra of the Cu⁺ peak was centered at 932.5 eV, whereas the Cu²⁺ peak of the binding energy was at 933.7 eV.^33,34 In addition, the spectra demonstrated clearly that increasing Na concentration offsets Cu peaks slightly in the direction of lower binding energy (Fig. 6), suggesting that a high Na concentration favors the formation of samples with a high Cu⁺/Cu²⁺ ratio. Our results indicated that Cu valence changed significantly and Cu⁺ was in a dominant position when the Na doping concentration exceeded 10%. Combined with our previous study,³⁴ we have found that the Cu⁺/Cu²⁺ ratio could be tuned by Na doping.

Fig. 2 SEM images of the Zn_0.95−xCu_0.05Na_xO nanocrystals prepared at (a) x = 0, (b) x = 0.005, (c) x = 0.01, (d) x = 0.02, (e) x = 0.05, (f) x = 0.1; and (g and h) TEM images of the Zn_0.9Cu_0.05Na_0.05O nanocrystals with different enlargements.

Fig. 3 XPS of the Zn 2p core level of the Zn_0.95−xCu_0.05Na_xO nanocrystals.

Fig. 4 XPS of the Na 1s core level for the Zn_0.95−xCu_0.05Na_xO nanocrystals.

Fig. 5 XPS of the O 1s core level and its corresponding Gaussian curve fittings for the Zn_0.95−xCu_0.05Na_xO nanocrystals.

Fig. 6 XPS of the Cu 2p core level and its corresponding Gaussian curve fittings for the Zn_0.95−xCu_0.05Na_xO nanocrystals.

EPR spectroscopy was employed to examine the paramagnetic characteristics of Zn_0.95−xCu_0.05Na_xO nanocrystals (Fig. 7(a)). The spectra revealed intrinsic defects of the gas-sensing nanocrystals. A strong EPR signal was observed centered at g = 2.0034 for Zn_0.95−xCu_0.05Na_xO, which could be ascribed to unpaired electrons trapped on oxygen vacancy sites. The stronger the EPR signal, the larger the concentration of oxygen vacancy defects. The EPR signal corresponds to the electrons trapped at surface O-vacancies, which facilitate the adsorption of oxygen molecules, reducing O₂ to ˙O₂. The signals of Zn_0.95−xCu_0.05Na_xO nanocrystals are stronger with increased Na doping, further confirming that the intrinsic defects of the samples increase with Na doping, in agreement with the XRD and XPS results. Considering these results together, we conclude that samples with a higher Cu⁺/Cu²⁺ ratio have more oxygen vacancy defects. A further understanding of the NO₂ gas-sensing mechanism and the influence of Na⁺ on Zn_0.95−xCu_0.05Na_xO can be achieved through the acquisition of information on photogenerated charge carrier separation and carrier transport direction using the surface photovoltage technique. We have discussed in a previous study³⁵ that ZnO is an n-type semiconductor with a positive SPV response due to photoinduced holes moving to the surface of ZnO. The stronger the SPV signal, the more photogenerated holes accumulate on the surface. That being said, the separation efficiency of photogenerated charges is higher with a higher SPV response, which will cause more O₂⁻ to react with photo-generated charges. It is noteworthy that samples containing higher Na doping and lower Cu²⁺/Cu⁺ ratio allow for higher SPV intensity, suggesting that large amounts of Cu⁺ and oxygen vacancies enhance the separation of photo-generated charges of the photoelectric gas sensor.

Fig. 7 (a) Room-temperature EPR spectra of the Zn_0.95−xCu_0.05Na_xO-based sensors. (b) Isothermal curves of the Zn_0.95−xCu_0.05Na_xO nanocrystals (x = 0, 0.01, 0.02, 0.05, and 0.1) during the nitrogen adsorption–desorption process. The inset figure shows pore size distribution plot of the Zn_0.95−xCu_0.05Na_xO nanocrystals (x = 0, 0.01, 0.05, and 0.1).

Fig. 7(b) presents the nitrogen adsorption–desorption isotherms for Zn_0.95−xCu_0.05Na_xO nanocrystals. For all samples, the isotherms show a gentle increase in the low relative pressure region (P/P_o < 0.1), while a distinct H3-type hysteresis loop³⁶ appears in the medium-to-high pressure range (P/P_o = 0.4–1.0), as seen in the inset of Fig. 7(b). The closure point of the hysteresis loop is located at P/P_o ≈ 0.42. Combined with the absence of a steep adsorption rise at low P/P_o, this indicates that the materials lack microporous structures but contain open slit-shaped mesopores formed by the stacking of rod-like nanocrystals,³⁷ which is consistent with the SEM observations in this study. Such stacked slit-like pore structures not only provide fast diffusion pathways for gas molecules but also offer abundant surface reaction sites, laying a structural foundation for excellent gas-sensing performance. The specific surface area was calculated using the Brunauer–Emmett–Teller (BET) theoretical model. The pore-size distribution was determined by analyzing the desorption branch data using the Barrett–Joyner–Halenda (BJH) model. The results indicate that Zn_0.95Cu_0.05O nanocrystals exhibit the highest specific surface area, approximately 5.4288 m² g⁻¹, while Zn_0.9Cu_0.05Na_0.05O nanocrystals exhibit only about 1.7684 m² g⁻¹. Alev et al.³⁸ reported that doping in ZnO can effectively regulate grain boundary migration and surface energy, thereby promoting the formation of nanostructures and the development of pore channels. In addition, Na⁺ may act as a flux during high-temperature synthesis, suppressing excessive grain growth and facilitating the preservation of more microporous structures. As evidenced by the data in Table 1, Zn_0.95Cu_0.05O possesses the highest specific surface area and the smallest average pore size. With increasing Na doping levels, the specific surface area of the samples initially decreased before exhibiting a slight recovery, while the average pore size showed a consistent upward trend. This behavior can be primarily ascribed to lattice distortion and altered grain growth dynamics induced by Na⁺ doping. Its incorporation leads to significant expansion of the ZnO lattice, generating numerous point defects and dislocations. Moreover, during high-temperature sintering, Na⁺ may act as a fluxing agent, lowering the activation energy for grain boundary migration and promoting the coalescence of smaller grains into larger ones. Consequently, the average particle size increases, accompanied by a reduction in specific surface area.

Table 1 Nitrogen adsorption–desorption isotherms for Zn_0.95−xCu_0.05Na_xO nanocrystals

Samples	Surface area/(m^2 g^−1)	Average pore sizes/nm	Pore volumes/(cm^3 g^−1)
Zn0.95Cu0.05O	5.4288 m^2 g^−1	12.2719 nm	0.017472 cm^3 g^−1
Zn0.94Cu0.05Na0.01O	3.7782 m^2 g^−1	16.1809 nm	0.015784 cm^3 g^−1
Zn0.9Cu0.05Na0.05O	1.7684 m^2 g^−1	16.9944 nm	0.006860 cm^3 g^−1
Zn0.85Cu0.05Na0.1O	2.1119 m^2 g^−1	20.5364 nm	0.006514 cm^3 g^−1

---

Combined with the gas-sensitive response test results in Fig. 8(a), although Zn₀.₉₅Cu₀.₀₅O exhibited the highest specific surface area, the Zn₀.₈₅Cu₀.₀₅Na₀.₁O sample showed the best response to 1-ppm NO₂ gas. This indicates that the gas-sensing performance depends not only on specific surface area but also on surface chemical activity and the distribution of surface oxygen species. Wang³⁹ suggested that appropriately enlarging pore size facilitates rapid diffusion of gas molecules within the material, thereby reducing the response time. Therefore, the excellent gas-sensing performance of Zn₀.₈₅Cu₀.₀₅Na₀.₁O could be attributed to its larger pore size (20.5364 nm), which promotes gas diffusion, along with Na doping that optimizes the surface electronic structure, synergistically improving the gas-sensing properties.

Fig. 8 (a) Real-time sensing responses of the NO₂ gas-sensing Zn_0.95−xCu_0.05Na_xO nanocrystals at room temperature (250 ppb, 500 ppb, 750 ppb and 1 ppm). (b) Repeatability of the Zn_0.9Cu_0.05Na_0.05O sensor response to 1.25-ppm NO₂ at room temperature. The sensing responses of the Zn_0.9Cu_0.05Na_0.05O sensor to 2-ppm NO₂ at (c) 11.3% relative humidity (RH), (d) 57.6% RH, and (e) 77.3% RH. (f) The sensing response curves of the Zn_0.9Cu_0.05Na_0.05O sensors to NO₂ at room temperature (50 ppb, 100 ppb, 150 ppb and 200 ppb). (g) Selectivity test of the Zn_0.9Cu_0.05Na_0.05O gas sensor for different gases.

3.2 Gas-sensing performance and enhancement mechanism

The NO₂ gas-sensing performances of Zn_0.95−xCu_0.05Na_xO-based sensors were investigated at room temperature. The comparative time dependence of the sensor response resistance shifts against 250 ppb, 500 ppb, 750 ppb, and 1 ppm NO₂ under 365-nm light irradiation is shown in Fig. 8(a). The fabricated sensors demonstrated excellent sensing performance at room temperature with relatively stable baselines, as shown in Table 2. Table 3 shows the NO₂ gas-sensing performance of different composite materials. The electrical resistance of the sensors enhances after exposure to NO₂ and then recovers immediately once exposed to dry air. The dynamic response curves indicate that the sensors coated with a sensitive material containing a high Cu⁺/Cu²⁺ ratio due to higher Na doping concentration show higher sensitivity towards the tested gas compared with those coated with a material with a low Cu⁺/Cu²⁺ ratio, as shown in SPV. Even at 250 ppb NO₂, the sensors coated with the sensitive material with a high Cu⁺/Cu²⁺ ratio showed stable response signals. Fig. 8(b) illustrates the repeatability of the Zn_0.9Cu_0.05Na_0.05O nanoparticle-based sensor response for an NO₂ gas concentration of 1.25 ppm at room temperature, where no obvious change in the amplitude of the sensing signal was seen over five cycles. Under relative humidity levels of 11.3%, 57.6%, and 77.3%, the Zn_0.9Cu0_.05Na_0.05O sensor exhibited relative humidity-dependent responses to 2 ppm NO₂, as shown in Fig. 8(c)–(e), respectively. The experimental results indicate that the fabricated Zn_0.9Cu_0.05Na_0.05O sensor exhibits excellent sensing performance, achieving a NO₂ detection limit of as low as 50 ppb, as shown in Fig. 8(f). In addition, the selectivity to NO₂ was confirmed by testing gases, including isopropanol, carbon monoxide, formaldehyde, ethanol, carbon dioxide, and hydrogen (Fig. 8(f)). The sensor fabricated using Zn_0.9Cu_0.05Na_0.05O exhibited a significantly stronger response to NO₂ gas at 0.25 ppm compared with the other interference gases at 20 ppm. The corresponding sensitivity values were 0.63, 0.2, 0.18, 0.174, 0.162, 0.12, and 0.1, respectively. These results indicate that the Zn_0.9Cu_0.05Na_0.05O sensor possesses favorable selectivity toward nitrogen dioxide.

Table 2 NO₂ gas-sensing responses of each sample at varying concentrations

Response sample	250 ppb	500 ppb	750 ppb	1 ppm
Zn0.95Cu0.05O	8%	13%	25%	45%
Zn0.945Cu0.05Na0.005O	7%	13%	30%	66%
Zn0.94Cu0.05Na0.01O	10%	16%	37%	68%
Zn0.93Cu0.05Na0.02O	18%	34%	67%	125%
Zn0.9Cu0.05Na0.05O	63%	76%	125%	182%
Zn0.85Cu0.05Na0.1O	68%	100%	153%	241%

---

Table 3 Nitrogen dioxide gas-sensing performance of different composite materials

Sensor materials	Gas concentration/ppm	Operation temperature/°C	Response	References
Ni–ZnO	10	250	120.8%	40
ZnO–Au	10	200	40%	41
Li–ZnO	30	25	70.25%	42
Fe–ZnO	100	400	31.81%	43
rGO-CuO/ZnO	100	25	77%	44
Pd–ZnO	2	100	46.3%	45
CuO–ZnO	10	140	9.7%	46
Zr–ZnO	0.5	150	13.63%	47
Li–Cu: ZnO	75	210	2415.26%	48
Cu–ZnO	1	200	3.7%	49
Pd@Pt-ZnO	0.05	80	60.3%	50
Ce–ZnO	100	250	8.6%	51
CuO–ZnO	100	25	36.7%	52
Ni–ZnO	20	25	264%	53
Zn0.9Cu0.05Na0.05O	0.25	25	63%	This work

---

From the above results, we conclude that Na doping enhances the NO₂ sensing performance of UV-activated sensors composed of Zn_0.95−xCu_0.05Na_xO nanocrystals. Further understanding of the role of Na ions in the NO₂ sensing of light-activated Zn_0.95−xCu_0.05Na_xO nanocrystals requires an understanding of the separation of photo-generated charge carriers and their transfer direction. Such an information can be investigated via surface photovoltage analysis conducted in an air atmosphere. Fig. 9 presents the maximum photovoltage response of the Zn_0.95−xCu_0.05Na_xO nanocrystals, which occurs at 350–370 nm. It is noteworthy that the sample with a higher Cu⁺/Cu²⁺ ratio exhibits a stronger surface photovoltage intensity, suggesting that Na doping increases the separation of photo-generated charges. The rapid transfer of charge carriers is extremely important for photoelectric gas sensors. In air, the holes generated by photon absorption integrate with the nanoparticle surface oxygen ions, leading to the desorption of oxygen species. Furthermore, O₂ captures the remaining electrons of separated electron–hole pairs to produce O₂⁻. To reach equilibrium between the adsorption and desorption of O₂, the electron depletion region beneath the sample surface changes at room temperature. Exposure to the stronger oxidizing NO₂ widens the depletion layers by increasing resistance or decreasing electric current. When NO₂ gas is removed, the electrons trapped in NO₂⁻ and O₂⁻ return to the surface of the nanocrystals to recombine with the holes, resulting in the narrowing of the depletion layer.

Fig. 9 Surface photovoltage (SPV) spectra of the Zn_0.9Cu_0.05Na_0.05O nanocrystals.

In fact, extending the lifetime of electron–hole pairs is significant for light-assisted room temperature NO₂ gas detection. As a consequence, the rapid recombination of photogenerated electrons and holes is inhibited by the synergistic effects of CuO decoration, oxygen vacancies, and Cu⁺/Cu²⁺ ratio in our experiments. ZnO/CuO heterojunction has been drawn according to the experimental data in Fig. 10. The light-assisted enhanced gas-sensing mechanism for our Zn_0.95−xCu_0.05Na_xO samples can be explained as follows: ZnO has a hexagonal wurtzite structure, which is a relatively open close-packed lattice. Na⁺ ions substitute for Zn in half of the tetrahedral sites, and more oxygen atoms are allowed to sit in the lattice sites, thereby generating oxygen vacancies. The substitution of Zn²⁺ with Na⁺ could introduce more defects and vacancies into the nanocrystal lattice. Surface oxygen vacancies can evoke a deeply trapped doubly charged oxygen vacancy (V_O˙˙), which can capture the electrons beside the conduction band of ZnO to inhibit the recombination of photogenerated electron holes. Surface oxygen vacancies are correlated with Na doping, as demonstrated by the SPV results. NO₂ adsorption of the samples, therefore, is increased with Na⁺ doping, thereby enhancing gas sensitivity. In ZnO, sodium (Na) substitutes for zinc (Zn) as a monovalent cation (Na⁺). The substitution of Na⁺ for the divalent Zn²⁺ introduces an effective negative charge, formally creating an acceptor defect denoted as Na_Zn. To maintain macroscopic charge neutrality, the system undergoes a charge compensation process, which in this case involves the reduction of a portion of the co-present copper ions from the Cu²⁺ to the Cu⁺ state. As the concentration of Na increases, the proportion of Cu⁺ ions rises concomitantly. These monovalent Cu⁺ ions subsequently occupy Zn²⁺ lattice sites, forming a defect center (Cu_Zn) that carries an effective single negative charge relative to the lattice. The accumulation of these negatively charged (Cu_Zn) centers necessitates further compensation to preserve the overall electrical neutrality of the crystal. A prevalent mechanism for this secondary compensation is the formation of intrinsic positively charged defects, specifically ionized oxygen vacancies (V_O˙˙). The formation of these donor-like vacancies effectively balances the negative charge from the acceptor defects, thereby stabilizing the doped lattice.

Fig. 10 Schematic of the gas-sensing mechanism of Zn_0.95−xCu_0.05Na_xO nanocrystals responding to NO₂ at room temperature.

On the other hand, Cu atoms are predicted to replace Zn within the ZnO lattice. The responses of the sensors are rapid. Zn_0.95−xCu_0.05Na_xO nanoparticle-based sensors with high Cu⁺/Cu²⁺ ratio demonstrate better gas-sensing responses compared with sensors with low Cu⁺/Cu²⁺ ratio. There is a direct correlation between gas concentration and electrical resistance change. Theoretically, the Zn_0.85Cu_0.05Na_0.1O sensor can detect lower concentrations than the measured minimum of 250-ppb NO₂. Cu²⁺ near the depletion layer and grain boundary can rapidly take up the photogenerated electrons to form Cu⁺ as follows:

Cu²⁺ + e⁻ = Cu⁺ (1)

This phenomenon leads to a wider depletion layer in Zn_0.95−xCu_0.05Na_xO nanoparticle-based sensors with a low Cu⁺/Cu²⁺ ratio (Fig. 10(b)). The variation of the Cu⁺/Cu²⁺ ratio upon electron attachment provides a continuous electron flux for the reduction reaction.^39,40 Based on the XPS results, excessive Na⁺ induces the Cu valence state to change from +2 to +1 when the Na concentration exceeds 10%. These chemical changes can be described as follows:

Cu⁺ + O₂ = Cu⁺ + O₂⁻ (2)

NO₂ + O₂⁻ = NO₂⁻+ O₂ (3)

From reactions eqn (1)–(3), Cu⁺ and Cu²⁺ evidently play a pivotal role in the absorption and desorption of a tested gas on the sensing material's surface.

Based on the experimental data in this study, we can draw a schematic diagram of the ZnO/CuO heterojunction energy band (Fig. 10(c)). The work function of p-CuO is larger than that of n-ZnO with low electronic emission possibility,⁴¹ where the conduction band offset (ΔE_c) of the p–n junction is about 0.12–0.3 eV. The generated free electrons in our samples could easily deliver the p–n junction due to the low ΔE_c, and simultaneously, the holes in CuO will amass close to the VB of CuO due to the large value of the barrier height (ΔE_v).⁴² We noticed that the electron mobility of CB will further improve because the ΔE_c value is lower than the ΔE_v value.^43,44 Correspondingly, the higher the separation rate of the photoexcited electrons and holes, the more the electrons and holes participate in redox reactions. We find that the ZnO/CuO heterojunctions reveal in all samples, which should not be the main cause of gas sensitivity changes in this work. In view of the preceding discussion, it is now possible to ascertain how Cu ion valence state changes contribute to the modulation of the depletion layer to endow the devices with gas-sensing capabilities. For Zn_0.95−xCu_0.05Na_xO nanoparticle-based sensors with higher Cu⁺/Cu²⁺ ratios, the generated Cu⁺ leads to a higher density of surface-accumulated O²⁻. That is, the presence of large amounts of Cu⁺ is beneficial for the efficient modulation of the depletion layer. The Cu⁺/Cu²⁺ ratio affects the coverage degree of O₂⁻, which directly influences NO₂ gas responsivity.^54–59

4. Conclusions

We herein propose a facile strategy to further enhance the gas-sensing performance of ZnO-based sensors. In this study, the efficient separation of electron–hole pairs is the main driving factor behind the enhancement of gas-sensing efficiency, resulting from the synergistic effects of high Cu⁺/Cu²⁺ ratio and surface oxygen vacancies. In particular, Zn_0.85Cu_0.05Na_0.1O-based sensors show excellent gas-sensing response against NO₂ at 250 ppb at room temperature. It is believed that the protocol can be devoted to designing other types of highly efficient ZnO-based sensors incorporating other metals cations with the valence shift, such as iron, manganese, cobalt, and vanadium.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the manuscript. For requests regarding the original data, please do not hesitate to contact the corresponding author.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant No. 61901401), the Open Project of National Laboratory of Solid State Microstructures, Nanjing University (Grant No. M38083) and the Fundamental Research Funds for the Central Universities, Southwest Minzu University (Grant No. ZYN2026107).

References

1. J. Park, R. Ghosh, M. S. Song, Y. J. Hwang, Y. B. Tchoe, R. K. Saroj, A. Ali, P. Guha, B. Kim, S. W. Kim and G. C. Yi, Individually addressable and flexible pressure sensor matrixes with ZnO nanotube arrays on graphene, NPG Asia Mater., 2022, 14, 40.
2. V. V. Volkov, D. J. Oliver and C. C. Perry, Polariton condensation and surface enhanced Raman in spherical ZnO microcrystals, Nat. Commun., 2020, 11, 4908.
3. X. H. Jiang, G. Liu, L. P. Tang, A. Z. Wang, Y. Tian, A. Wang and Z. L. Du, Quantum dot light-emitting diodes with an Al-doped ZnO anode, Nanotechnology, 2020, 31, 255203.
4. W. X. Ouyang, J. X. Chen, Z. F. Shi and X. S. Fang, Self-powered UV photodetectors based on ZnO nanomaterials, Appl. Phys. Rev., 2021, 8, 031315.
5. J. Bang, Y. S. Kim, C. H. Park, F. Gao and S. B. Zhang, Understanding the presence of vacancy clusters in ZnO from a kinetic perspective, Appl. Phys. Lett., 2014, 104, 252101.
6. M. J. Zheng, P. B. Gui, X. Wang, G. Z. Zhang, J. X. Wan, H. Zhang, G. J. Fang, H. Wu, Q. Q. Lin and C. Liu, ZnO ultraviolet photodetectors with an extremely high detectivity and short response time, Appl. Surf. Sci., 2019, 481, 437–442.
7. A. S. Bhosale, K. K. Abitkar, P. S. Sadalage, K. D. Pawar and K. M. Garadkar, Photocatalytic and antibacterial activities of ZnO nanoparticles synthesized by chemical method, J. Mater. Sci.:Mater. Electron., 2021, 32, 20510–20524.
8. S. J. Jiang, K. L. Lin and M. Cai, ZnO Nanomaterials: Current Advancements in Antibacterial Mechanisms and Applications, Front. Chem., 2020, 8, 580.
9. Z. L. Li, Z. Q. Li, C. L. Zuo and X. S. Fang, Application of nanostructured TiO₂ in UV photodetectors: a review, Adv. Mater., 2022, 34, 2109083.
10. D. Banerjee, I. M. Asuo, A. Pignolet, R. Nechache and S. G. Cloutier, High performance photodetectors using porous silicon-TiO₂ heterostructure, Eng. Res. Express, 2020, 2, 035021.
11. M. X. Ge, D. Q. Xie, C. Jiao, Y. W. Yang, L. Shen, M. B. Qiu, H. Zhang, Z. J. He, H. X. Liang and Z. J. Tian, Mechanical properties and biocompatibility of MgO/Ca₃(PO₄)₂ composite ceramic scaffold with high MgO content based on digital light processing, Ceram. Int., 2022, 48, 21175–21186.
12. S. V. Gudkov, D. E. Burmistrov, D. A. Serov, M. B. Rebezov, A. A. Semenova and A. B. Lisitsyn, A Mini Review of Antibacterial Properties of ZnO Nanoparticles, Front. Phys., 2021, 64, 1481.
13. S. Sulaiman, S. Izman, M. B. Uday and M. F. Omar, Review on grain size effects on thermal conductivity in ZnO thermoelectric materials, RSC Adv., 2022, 12, 5428–5438.
14. F. Giovannelli, C. Chen, P. Díaz-Chao, E. Guilmeau and F. Delorme, Thermal conductivity and stability of Al-doped ZnO nanostructured ceramics, J. Eur. Ceram. Soc., 2018, 38, 5015–5020.
15. K. Kaviyarasu, T. Khamlich, C. M. Magdalane and M. Maazaab, Stability and thermal conductivity of CuO nanowire for catalytic applications, J. Environ. Chem. Eng., 2019, 7, 103255.
16. T. Eisenmann, J. Asenbauer, S. J. Rezvani, T. Diemant, R. J. Behm, D. Geiger, U. Kaiser, S. Passerini and D. Bresser, Impact of the transition metal dopant in zinc oxide lithium-ion anodes on the solid electrolyte interphase formation, Small Methods, 2021, 5, 2001021.
17. S. Soltani, K. Akhbari and J. White, Synthesis, crystal structure, magnetic, photoluminescence and antibacterial properties of dinuclear Copper (II) complex, J. Mol. Struct., 2020, 1214, 128233.
18. S. Roguai and A. Djelloul, A structural and optical properties of Cu-doped ZnO films prepared by spray pyrolysis, Appl. Phys. A, 2020, 126, 122.
19. Q. Ma, X. Yang, X. Lv, H. Jia and Y. Wang, Cu doped ZnO hierarchical nanostructures: morphological evolution and photocatalytic property, J. Mater. Sci.: Mater. Electron., 2019, 30, 2309–2315.
20. S. Brahma, Y. W. Yeh, J. L. Huang and C. P. Liu, Cu-doped p-type ZnO nanostructures as unique acetone sensor at room temperature (∼25 °C), Appl. Surf. Sci., 2021, 564, 150351.
21. R. S. Ganesh, E. Durgadevi, M. Navaneethan, V. L. Patil, S. Ponnusamy, C. Muthamizhchelvan, S. Kawasaki, P. S. Patil and Y. Hayakawa, Tuning the selectivity of NH₃ gas sensing response using Cu-doped ZnO nanostructures, Sens. Actuators, A, 2018, 269, 331–341.
22. C. Wu and Q. Huang, Synthesis of Na-doped ZnO nanowires and their photocatalytic properties, J. Lumin., 2010, 130, 2136–2141.
23. A. Sáaedi and R. Yousefi, Improvement of gas-sensing performance of ZnO nanorods by group-I elements doping, J. Appl. Phys., 2017, 122, 224505.
24. M. H. Elsayed, T. M. Elmorsi, A. M. Abuelela, A. E. Hassan, A. Z. Alhakemy, M. F. Bakr and H. H. Chou, Direct sunlight-active Na-doped ZnO photocatalyst for the mineralization of organic pollutants at different pH mediums, J. Taiwan Inst. Chem. Eng., 2020, 115, 187–197.
25. R. Jaisutti, M. Y. Lee, J. Kim, S. Choi, T. J. HA, J. Kim, H. Kim, S. K. Park and Y. H. Kim, Ultrasensitive room-temperature operable gas sensors using p-type Na:ZnO nanoflowers for diabetes detection, ACS Appl. Mater. Interfaces, 2017, 9, 8796–8804.
26. H. Yuan, M. Xu, C. J. Dong, J. Ma and X. Y. Wang, Mechanistic insights into magnetic and gas sensing properties of (F, Na)-codoped ZnO nanocrystals by room-temperature, Appl. Surf. Sci., 2019, 496, 143511.
27. Q. P. Zhang, Z. M. Pang, W. Y. Hu, J. Li, Y. T. Liu, F. Yu, C. W. Zhang and M. Xu, Performance degradation mechanism of the light-activated room temperature NO₂ gas sensor based on Ag-ZnO nanoparticles, Appl. Surf. Sci., 2021, 541, 148418.
28. Q. P. Zhang, G. Z. Xie, M. Xu, Y. J. Su, H. L. Tai, H. F. Du and Y. D. Jiang, Visible light-assisted room temperature gas sensing with ZnO-Ag heterostructure nanoparticles, Sens. Actuators, B, 2017, 259, 269–281.
29. M. Xu, H. Yuan, B. You, P. F. Zhou, C. J. Dong and M. Y. Duan, Structural, optical, and magnetic properties of (Co, Cu)-codoped ZnO films with different Co concentrations, J. Appl. Phys., 2014, 115, 093503.
30. J. L. Shen, S. Jiang, Y. C. Xu, M. Y. Li, S. Zhu, Z. B. Chen, X. F. Lin, H. Liu, H. L. Li and J. Z. Zhang, Boron and sodium co-doped ZnO varistor with high stability of pulse current surge, J. Alloys Compd., 2017, 728, 368–375.
31. K. Ganapathi, M. Kaur, M. Pathak, M. Shaheera, A. Pathak, S. C. Gadkari and A. K. Debnath, Highly sensitive NO₂ sensor based on ZnO nanostructured thin film prepared by SILAR technique, Sens. Actuators, B, 2021, 335, 129678.
32. Z. Han, L. Ren, Z. Cui, C. Chen, H. Pan and J. Chen, Ag/ZnO flower heterostructures as a visible-light driven photocatalyst via surface plasmon resonance, Appl. Catal., B, 2012, 126, 298–305.
33. A. Vij, S. Gautam, S. O. Won, A. Thakur, I. J. Lee and K. H. Chae, X-ray photoelectron spectroscopy of Zn0.98Cu0.02O thin film grown on ZnO seed layer by RF sputtering, Mater. Lett., 2012, 88, 51–53.
34. H. Yuan, M. Xu and X. S. Du, Effects of Co doping on the structural and optical properties of ZnCuO thin films, Mater. Lett., 2015, 154, 94–97.
35. J. Y. Wang, J. H. Deng, Y. B. Li, H. Yuan and M. Xu, ZnO nanocrystal-coated MoS₂ nanosheets with enhanced ultraviolet light gas sensitive activity studied by surface photovoltage technique, Ceram. Int., 2020, 46, 11427–11431.
36. M. Thommes, K. Kaneko, A. V. Neimark, J. P. Olivier, F. Rodriguez-Reinoso, J. Rouquerol and K. S. W. Sing, Physisorption of gases, with special reference to the evaluation of surface area and pore size distribution, Pure Appl. Chem., 2015, 87, 1051–10269.
37. J. Zhang, S. R. Wang, M. J. Xu, Y. Wang, B. L. Zhu, S. M. Zhang, W. P. Huang and S. H. Wu, Hierarchical porous ZnO architectures for gas sensing application, CrystEngComm, 2013, 15, 5966–5973.
38. O. Alev, İ. Ergün, O. Özdemir, L. Çolakerol Arslan, S. Büyükköse and Z. Z. Öztürk, Enhanced ethanol sensing performance of Cu-doped ZnO nanorods, Mater. Sci. Semicond. Process., 2021, 136, 106149.
39. J. Wang, C. Y. Hu, Y. Xia and B. Zhang, Mesoporous ZnO nanostructures with highly exposed (0001) facets for ultra-sensitive gas sensors, J. Mater. Chem. A, 2020, 8, 7978–7987.
40. W. Oum, A. Mirzaei and Y. K. Shin, et al., Enhancement of NO₂ sensing of ZnO by irradiating with high-energy Ni ions, Sens. Actuators, B, 2025, 435, 137615.
41. A. Pathak, S. Samanta and J. Bahadur, et al., Tailoring ZnO-Au nano-composites for enhanced NO₂ sensing: Synergistic effect of facet orientation and Au nano-particle surface functionalization, Appl. Surf. Sci., 2025, 687, 162199.
42. N. Singh, N. Singh and A. Bhargav, et al., Li-substituted ZnO nanoparticles exhibiting room temperature optical gas sensing for NO₂ with swift response and recovery, Opt. Quant. Electron., 2024, 57, 62.
43. V. M. Hiremath, N. Momin and V. M. Kangralkar, et al., Synthesis and characterization of Fe-doped ZnO films for enhanced NO₂ gas-sensing applications, J. Korean Phys. Soc., 2024, 1–11.
44. D. Yadav, S. P. Shukla and D. G. Varma, Synthesis of rGO-CuO/ZnO nanocomposites for humidity tolerant room temperature NO₂ gas sensor, Appl. Phys. A, 2024, 130, 713.
45. M. H. Luu, T. T. T. Pham and D. V. Nguyen, et al., Excellent NO₂ sensor based on porous Pd-ZnO nanorods prepared by a facile hydrothermal method, Adv. Nat. Sci. Nanosci. Nanotechnol., 2024, 15, 035005.
46. S. Li, L. Yu and C. Zhang, et al., Controllable synthesis of heterostructured CuO–ZnO microspheres for NO₂ gas sensors, Sens. Actuators, B, 2024, 417, 136179.
47. J. Rodrigues, S. Tripathy and G. N. Shimpi, Enhanced performance of 1D rod shaped Zirconium (Zr) decorated over spherical shaped Zinc oxide (ZnO) nanostructures for NO₂ gas, Opt. Mater., 2024, 152, 115516.
48. K. K. Jasmi, A. T. Johny and S. V. Siril, et al., Influence of lithium on Cu-doped ZnO thin films fabricated via sol-gel spin coating technique for improved NO₂ gas sensing applications, J. Electroceram., 2024, 1–10.
49. A. Umar, A. A. Ibrahim and N. A. Begi, et al., Synthesis of Bitter gourd-shaped Cu-doped ZnO nanostructures and their investigation for the detection of NO₂ gas at low concentrations, Ceram. Int., 2024, 50, 11320–11328.
50. C. Sui, M. Zhang and Y. Li, et al., Pd@Pt Core-Shell Nanocrystal-Decorated ZnO Nanosheets for ppt-Level NO₂ Detection, ACS Sens., 2024, 1967–1977.
51. U. Ahmad, A. Sheikh and K. Rajesh, et al., Ce-doped ZnO nanostructures: A promising platform for NO₂ gas sensing, Chemosphere, 2023, 349, 140838.
52. S. Smriti, D. Rita and R. Suman, et al., The development of CuO-ZnO based heterojunction for detection of NO₂ gas at room temperature, Appl. Phys. A, 2023, 129, 10.
53. Y. H. Feng, W. Hong and L. Di, et al., Pompon-like Ni doped ZnO mesoporous microspheres for conductometric NO₂ sensing properties at room temperature, Sens. Actuators, B, 2023, 394, 134388.
54. X. Y. Chen and D. H. Kuo, Nanoflower bimetal CuInOS oxysulfide catalyst for the reduction of Cr (VI) in the dark, ACS Sustain. Chem. Eng., 2017, 5, 4133–4143.
55. S. C. Riha, D. C. Johnson and A. Y. Prieto, Cu₂Se nanoparticles with tunable electronic properties due to a controlled solid-state phase transition driven by copper oxidation and cationic conduction, J. Am. Chem. Soc., 2011, 133, 1383–1390.
56. S. Park, H. Ko, S. Kim and C. Lee, Role of the Interfaces in multiple networked one-dimensional core-shell nanostructured gas sensors, ACS Appl. Mater. Interfaces, 2014, 6, 9595–9600.
57. C. Wang, X. Q. Fu, X. Y. Xue, Y. G. Wang and T. H. Wang, Surface accumulation conduction controlled sensing characteristic of p-type CuO nanorods induced by oxygen adsorption, Nanotechnology, 2007, 18, 145506.
58. M. Thirumoorthi, S. ShekDhavu, V. Ganesh, T. H. Abdulaal, I. S. Yahia and D. Deivatamil, High responsivity n-ZnO/p-CuO heterojunction thin film synthesised by low-cost SILAR method for photodiode applications, Opt. Mater., 2022, 128, 112410.
59. P. P. Subha and M. K. Jayaraj, Enhanced room temperature gas sensing properties of low temperature solution processed ZnO/CuO heterojunction, BMC Chem., 2019, 13, 4.

---

Footnote

† These authors contributed equally to this work.

---