Synthesis of Cu₂PO₄OH

Details about chemicals and materials can be found in the Supplementary Material (SM). Appropriate amounts of Cu(NO₃)₂·2.5 H₂O and NH₄H₂PO₄ (ratio 2:1) were dissolved separately in deionized H₂O and stirred for 30 min. The phosphate solution was slowly added to the copper solution and left to stir vigorously. NH₃ solution was added dropwise until pH was 6.9. Afterward, samples were aged under different conditions. Sample Lib01 was transferred into a Teflon-lined stainless-steel autoclave of 125 mL capacity and heated at 150°C for 6 h. Sample Lib02 was also aged using the same autoclave and was heated at 180°C for 6 h. Sample Lib03 was aged in a 250 mL round-bottom flask attached to a condenser and heated at 60°C for 24 h with vigorous stirring. Sample Lib04 was left to stir in a beaker at room temperature for 16 h. Afterward, the precipitates were recovered and washed several times by centrifugation at 4750 rpm multiple times before being dried in an oven at 80°C overnight.

Sample Lib04 was additionally calcined in a furnace at 400°C for 2 h at 2°/min in the air. This sample is labeled Lib05. Sample Lib04 was easily scaled-up in the laboratory facilities, adapting the experimental procedure mentioned above with larger equipment and glassware. The upscaled sample was chosen for the application trials after calcination at 400°C.

IR security ink tests

Appropriate amounts of the IR-transparent ink and libethenite pigment were introduced in a disposable cup together with three high-density alumina 12 mm beads. The contents were mixed at 1200 rpm for 5 min in a Speedmixer (DAC 150 FVZ, Hauschild). This operation was repeated several times to ensure proper milling, dispersion grade, and good ink consistency. The resulting inks were coated on cardboard using an Orange Proofer (IGT Testing Systems Pte. Ltd.) and samples were left to dry overnight at room temperature. The optical properties of the coated inks were checked with a ultraviolet-visible- NIR (UV-VIS-NIR) spectrophotometer.

Laser-marking tests

Low-density polyethylene plates were produced by injection molding (Hampton Colours Ltd., UK) by dispersing 1% of the additive in the polymer composition. Soft polyvinyl chloride (PVC) plates were produced in-house starting from the precursor soft PVC formulation. An appropriate amount of the formulation was mixed with the additive at 1200 rpm for 10 min in the Speedmixer. Afterward, the mixture was coated on a metallic surface to form a film. The film was cured at 115°C for 25 min in an oven. Finally, the film was left to cool down before being removed from the metallic surface by using a palette knife. All tests plates were laser marked (Nd:YVO₄ laser) using a lab-size laser-marking instrument (Zeichenlasermaschine Photonenwerke, Trumpf). Testing plates were evaluated visually and by the use of a microscope (GX microscope, CAM H5).

Characterization

The crystal structure of the powder samples was analyzed by x-ray diffraction using Cu K_α radiation (Bruker D8 Advance diffractometer). Infrared (IR) spectra were recorded on a Nicolet iS10 from ThermoScientific with Smart iTX (ATR). Raman analysis was performed on a DXR 2 Raman Microscope (Thermofisher). BET surface areas were calculated by using the Brunnauer, Emmet and Teller method on the adsorption data recorded on a Micromeritics TriStar II Surface area and Porosity analyzer. Particle morphology was analyzed with a scanning electron microscope (SEM) Hitachi S4500. SEM-energy dispersive x-ray (EDX) analysis was done with a TM3000 microscope (Hitachi). The optic properties of the samples were checked by a UV/VIS/NIR Spectrometer Lambda 750S from Perkin Elmer with a 100 mm Int. Sphere.

Characterization of Cu₂PO₄OH

All Cu₂PO₄OH samples are summarized in Table I. Figure (1a) shows the x-ray powder diffractograms of libethenite samples synthesized at pH 6.9 with variations in their aging treatments. It was possible to identify the orthorhombic Cu₂PO₄OH phase in all samples (JCPDS 83-2264). Samples Lib01 and Lib02 that were aged using an autoclave at 150 and 180°C, respectively, showed better-defined peaks with lower values of full width at half maximum (FWHM) compared with the rest of the samples (Table I). The effect of autogenous pressure on the crystallinity of the samples and crystal growth is well known, which was observed for samples Lib01 and Lib02. It was also possible to observe the appearance of an extra minor phase believed to be a copper phosphate precipitate (JCPDS 22-1038), but the low intensity of the peaks hindered phase assignation. The reason for the appearance of this extra copper phosphate phase could be related to different precipitation rates of the copper species. The values for crystallite sizes were calculated for all samples (Table I) using Scherrer's equation and the (220) peak.^[Reference Klug and Alexander^12] In general, larger values of crystallite size belong to samples aged with the autogenous pressure that favors crystal growth. Information regarding cell parameters can be found in the SM. When additives are added into polymers for Laser sensitive applications, they need to be stable in terms of chemical structure and composition above the temperatures where the plastic is being processed. Although temperatures may vary depending on the nature of the plastic,^[13] in general, additives should have good thermal stability up to 300°C.^[14, Reference Vasile^15] The thermal stability of Cu₂PO₄OH was checked by high-temperature x-ray powder diffraction (XRD) by heating a Lib03 sample from room temperature (RT) to 505°C and measuring the diffractograms in-situ every 30°C [Fig. (1b)]. In general, no significant changes were observed in the diffractograms as the temperature was raised. The shape and position of the peaks remained unchanged and no extra phases were detected during the heating process. The thermal stability of Cu₂PO₄OH observed in the high-temperature diffractogram confirms that copper hydroxyphosphate can be dispersed in a wide range of polymers for laser-marking purposes without suffering degradation.

Figure 1.

(a) Diffractogram of (a) Lib01, (b) Lib02, (c) Lib03, (d) Lib04, (e) Lib05 samples. (b) High-temperature diffractogram of Lib03 sample, recorded from RT until 505°C.

Table I.

Summary of Cu₂PO₄OH samples and XRD crystallographic data

^a Calculated by Scherrer's equation.

The morphology of the samples was checked by SEM [Fig. (2)] and particle sizes determined.^[Reference Schneider, Rasband and Eliceiri^16] In general, the morphology of Cu₂PO₄OH samples exhibited agglomerated particles forming spheres, which is expected when using a pH close to 7 during the synthesis. The size of these aggregated spheres differed greatly from the aging conditions of each sample.^[Reference Cho, Kim, Lee, Kwak, Bae, Noh, Yoon, Jung, Kim and Hong^10, Reference Xu, Jiao and Chen^17] Samples Lib01 and Lib02 [Figs. 2(a) and 2(b), respectively], presented walnut-shaped spheres with some differences. While sample Lib01 displayed an agglomeration where sheets could be observed in the walnut-shape spheres, sample Lib02 exhibited edges forming a cross and angular shapes within the spheres. Sizes for these spheres ranged from 13 to 21 µm for sample Lib01 and from 7 to 20 µm for sample Lib02. Cotton-like sphere morphology was obtained for samples Lib03, Lib04, and Lib05 [Figs. 2(c), 2(d) and 2(e), respectively]. The size of these spheres decreased dramatically for sample Lib03 compared with samples aged using an autoclave. It was observed that these spheres were formed by an agglomeration of small rounded particles with sizes ranging from 0.6 to 1.6 µm. Samples Lib04 and Lib05 showed spheres of different sizes that ranged between 4 and 11 µm and 4 and 10 µm, respectively. It is possible to observe a slight decrease in the size of these spheres for sample Lib05 compared with Lib04 sample, after calcination at 400°C. During the synthesis, the pH and [Cu²⁺]/[PO₄³⁻] ratio was kept constant for all samples. Differences during crystallization and crystal growth are therefore attributed to temperature, pressure, and time applied during the synthesis. By changing these conditions, differences in the architecture of copper hydroxyphosphates can be observed.^[Reference Rudolph^18] While the mechanism of Cu₂PO₄OH has been established and it is clear that the morphology is strongly dependent on the concentration of NH₄⁺ ions as well as on the Cu²⁺ concentration and on the type of solvents, the role of temperature is much less studied. We believe that heating is normally used to avoid the formation of the well-known copper complex [Cu(NH₃)₄²⁺],^[Reference Cho, Kim, Lee, Kwak, Bae, Noh, Yoon, Jung, Kim and Hong^10] but interestingly this complex was not observed in sample Lib04, synthesized at RT. When this sample was upscaled and the concentration of the starting reagents increased to reduce the use of water, we could then observe the formation of a small dark blue supernatant liquid, which was discharged when washing the solid. When the crystal structure of the upscaled sample was checked no other phase was detected. An increase in the concentration led to spheres with a smaller size compared with sample Lib05, with sizes that ranged from 0.8 to 2 µm. Additional characterization for this upscaled sample can be found in the SM.

Figure 2.

SEM micrographs of (a) Lib01 (×3k), (b) Lib02 (×3k), (c) Lib03 (×3k), (d) Lib04 (×3k) and (e) Lib05 (×4k) samples, (f) reflectance spectra of libethenite samples.

The optical properties were checked by the measurement of the UV-Vis-NIR diffuse reflectance spectrum of the powdered samples [Fig. (2f)]. A very wide reflection curve centered at around 1125 nm appears in all samples, which is derived from the crystal structure of Cu₂PO₄OH and the Cu coordination. Differences in the reflectance spectra intensities can be explained by differences in the morphology of the samples and cell parameters, whose values can be correlated with the energy band gap.^[Reference Cho, Kim, Lee, Kwak, Bae, Noh, Yoon, Jung, Kim and Hong^10] Photoabsorption was higher for samples (Lib01 and Lib02) with walnut-shaped sphere morphology. Information regarding the origin of the NIR absorption based on density functional calculations for this compound has been previously reported.^[Reference Li, Dai, Ma, Zhu and Huang^19] The wide band that appears between 500 and 2000 nm in all samples is a superposition of different minor bands. Complete band analysis of this deconvoluted wide band can be found elsewhere according to the particular Cu(II) symmetry in Cu₂PO₄OH.^[Reference Gangi Reddy, Ramasubba Reddy, Siva Reddy, Lakshmi Reddy and Jagannatha Reddy^20] Details about FTIR, Raman, N₂ physisorption, and SEM-EDX characterization can be found in the SM.

The synthesis conditions of sample Lib04, which did not use any type of heating and successfully yielded the Cu₂PO₄OH phase, are very interesting from an industrial point of view. Unlike other reported syntheses that use autoclaves,^[Reference Rudolph^18, Reference Xiao, Sun, Meng, Yu, Yuan, Xu, Song, Jiang and Xu^21, Reference Zhan, Li and Chen^22] organic reagents^[Reference Xu, Zhang and Liu^23, Reference Fu, Wang, Wu, Wang, Wang, Wang, Zhang and Qiu^24] and/or surfactants,^[Reference Han, Li, Xu, Yuan, Wang and Yu^25] this simplified synthesis is much more energy efficient and it also avoids the cost of pressurized equipment. For these reasons, we successfully upscaled sample Lib04 to 300 g using standard laboratory equipment. To the best of our knowledge, the synthesis of libethenite at RT has never been reported before.

IR security ink trials

Banknotes incorporate numerous security features such as 3D images, intaglio printing, holograms, several types of inks, UV, and magnetic features, and logos with rolling color effects, in order to prevent counterfeiting. Banknotes incorporate IR inks which either absorb or reflect IR radiation and are identifiable by standard banknote detectors. In order to test the performance of libethenite as security additive, we blended copper hydroxyphosphate in an additive-free standard ink used in banknotes. The ink is transparent in the region between 700 and 1130 nm. Printing tests were performed by varying the percentage of additive incorporated in the range 0–20% [Fig. (3a)]. A blank was printed and the spectrum showed, in effect, no bands in the region between 700 and 1130 nm [Fig. (3b)]. It is difficult to assign ink bands, probably overlapped, in the region between 1200 and 2500 nm due to the fact that the exact composition of the ink is unknown to us. All tests with the additive in the formulation resulted in prints which showed a strong absorption in the region 630–1400 nm [Fig. (3b)], in agreement with the additive-only spectrum [Fig. S5(d)]. It is possible to correlate the increase of the additive in the composition with the increase in the band signal centered at 1140 nm. Ideally, a good additive for IR security inks should have a good performance when being used in low concentration in the ink formulation, in quantities of 10% or less. In addition to this, the CIE coordinates are also significant for the application, as additives with light colors are preferred over dark ones. We have calculated the L* a* b* coordinates with the spectrophotometer (Table II and SM) and compared the performance of this additive with a commercial product, StanoStat CPM10C [Figs. 3(c) and 3(d)]. While the spectra show that the NIR absorption of CPM10C is slightly better at the same percentages compared with the Cu₂PO₄OH additive, there are significant differences in the coloration of the test prints, as observed by the L* a* b* values for this material. Notorious differences in L* and b* parameters indicate that the color is dark blue, in comparison with a light green color for printing tests of copper hydroxyphosphate. The broad band between 700 and 1130 nm for this compound can function as a fingerprint band which can be easily detected by available commercial detectors. In addition, the L* a* b* coordinates exhibit desirable values for IR security inks, therefore suggesting the potential of this additive as a security feature in banknotes.

Figure 3.

(a) Print tests of the security IR ink with Cu₂PO₄OH (0%, 5%, 10%, 15%, and 20%), (b) reflectance spectra of print tests of the security IR ink with Cu₂PO₄OH, (c) print tests of the security IR ink with StanoStat CPM10C (0%, 5%, 10%, 15%, and 20%), (d) reflectance spectra of print tests of the security IR ink with StanoStat CPM10C.

Table II.

CIE L* a* b* coordinates

Recent findings about antibacterial properties of libethenite and other copper compounds^[Reference Zhao, Pang, Li, Li, Yan, Ma, Li, Chen, Zhang and Zheng^26–Reference Zhang, Tang, Li and Tang^28] could open the way towards the incorporation of Cu₂PO₄OH as an additive in antibacterial inks for packaging purposes.^[Reference Mahalik and Nambiar^29, Reference Cloutier, Mantovani and Rosei^30]

Laser-marking tests

The first set of test plates were manufactured by injection molding with low-density polyethylene (LDPE) as a polymer, while a second set were made with polyvinyl-chloride (PVC) as a polymer. In both sets, the additive concentration was kept at 1%. Commonly, the evaluation of an additive for laser-marking is assessed by marking a matrix of squares with varying line pitch (mm), marking speed (mm/s) and pulse frequency (kHz). These laser-marked squares are first visually evaluated to check the contrast, quality of mark and legibility (for laser-marked text), mark depth and surface defects. Afterward, microscopy characterization complements ocular evaluation. However, mark quality is ultimately assessed in compliance with the application requirements. In a typical test, a first matrix was laser-marked with the following value intervals: line pitch (0.03–0.09 mm), frequency (15–80 kHz) and marking speed (450–5000 mm/s). A second matrix was laser-marked choosing the appropriate range for each parameter that led to a good mark after an ocular evaluation and these were characterized using microscopy. In general, squares with good contrast were visually observed in all marked testing plates. Matrix LDPE1 [Fig. (4a)] shows squares with good contrast with line pitch values ranging from 0.03 to 0.05 mm and marking speed values ranging from 450 to 2142 mm/s. A second and optimized matrix [Fig. (4b)], LDPE2, was laser-marked using these intervals to produce a set of squares with good contrast. In order to understand the surface of the marked squares the quality of the spots was checked by microscopy [Fig. 4(c)]. Squares A2 and B2 exhibited a uniform color, however, it was possible to distinguish dots in square B2, which was marked with the same frequency (33 kHz) but with a larger value of line pitching (0.05 in contrast to 0.03 mm, for squares B2 and A2). In contrast, square B6 which was marked with the same line pitching value of 0.05 mm as square B2, but with a frequency of 70 kHz shows a less homogenous surface where the LDPE polymer appears partially marked. An increase in line pitching led to a less homogeneous mark, which can be observed in square D3, which was marked with a line pitching of 0.09 mm and 42 kHz of frequency. Using any combination of the conditions that were used for marking the squares in matrix LDPE2 it was possible to mark proper figures that exhibited a uniform and homogenous mark with sharp edges (SM).

Figure 4.

(a) Testing plate LDPE1, (b) testing plate LDPE2, (c) micrographs of four marked squares in testing plate LDPE1 (×100).

Markability depends as well on the type of polymer. Not all polymers may be marked by using the same conditions and type of laser. This feature can be clearly observed with PVC testing plates [Fig. (5)]. The first laser-marked matrix on PVC [Fig. (5a)], PVC1, shows different marking results using the same marking conditions as in matrix LDPE1. For this plate, best marking resolution was observed using line pitching of 0.03 mm and frequencies between 60–80 kHz and using line pitching of 0.05 mm and frequencies between 15 and 52 kHz. A second optimized matrix was laser-marked [Fig. (5b)], PVC2, by using a line pitching ranging between 0.03 and 0.05 mm and frequency between 33 and 80 kHz. Most of the squares exhibited visible and uniform marks [Fig. (5c)] with squares A5 and B2 exhibiting a uniform marked surface. By slightly altering the line pitching and frequency values important differences arose on the marked surface, as can be observed in square C3, which was marked with a line pitching of 0.05 mm and a frequency of 42 kHz compared with square B2 which was marked with a line pitching of 0.03 mm and a frequency of 33 kHz. Dots were not overlapped in square C3 and as a result, it exhibited a non-homogeneous mark. Similarly to matrix LDPE1, when using 0.09 mm of line pitching and a frequency of 80 kHz the resulting square was poorly marked (square D7). Thus, when the marking speed is too high a non-continuous marked line is produced and the mark appears as a series of spots. Based on our experiments, for both types of testing plates, a line pitching lower than 0.05 mm is recommended to ensure a proper mark. Pulse frequency is one of the most important parameters to be set up and depends on the type of polymer. For copper hydroxyphosphate as an additive, a frequency between 15 and 42 kHz for LDPE and between 33 and 80 kHz for PVC should be sufficient.

Figure 5.

(a) Testing plate PVC1, (b) testing plate PVC2, (c) micrographs of four marked squares in testing plate PVC1 (×100).

Ideal marks will show no scattering and good resolution on the edges in the figures and scattering was not observed in any of the test plates.^[Reference Han and Gubencu^31] Several figures and barcodes were additionally laser-marked, whose edges were sharp and uniform and this feature is recommended especially for bar- and QR codes which need to be easy to read by code scanners/readers. The mechanism that we propose for laser-marking for these two types of testing plates starts by directing the laser beam (1064 nm) onto the polymer's surface where copper hydroxyphosphate is dispersed. The additive readily absorbs the energy and a surface reaction starts, which could be a combination of melting through a thermochemical reaction and carbonization of the polymer.^[32, Reference Zhong, Cao, Qiu, Wu, Liu, Li and Zhu^33] During this reaction, the temperature of the irradiated area and its surroundings increases and polymer degradation leads to gas formation. Vapor residues were detected, which were extracted by the fume extraction unit inside the marking chamber and a minor quantity of polymer dust appeared on the surface, which was easily removed.