Introduction

The first article devoted to methods of synthesis and properties of colloidal gold was published by Michael Faraday as early as 1857. Since then, researches on metal nanoparticles (NPs) and their various applications are of increasing interest, as evidenced by the significant growth of scientific publications in this field (Zhang et al. 2016; Gawande et al. 2016; Amendola et al. 2017; Lim et al. 2015; Sirelkhatim et al. 2015; Kuznetsov et al. 2016).

The main efforts of modern researchers are aimed to synthesis of NPs with various sizes, shapes, and narrow size distributions, to search for new substances that stabilize them, to reveal the relationship between the size, shape, and properties of NPs and used reducing agents, stabilizers, and formation conditions, as well as further expansion of the NPs application (Apyari et al. 2014; Terenteva et al. 2017; Das et al. 2018; Goud et al. 2018; Malik and Mukherjee 2018; Xiao et al. 2018; Georgakilas et al. 2016; Zhu et al. 2015; Huang et al. 2017; Kaushik and Moores 2016).

Special attention is paid to the creation of new nanocomposite and hybrid materials. Inclusion of NPs into various solid matrices makes it possible to create new nanocomposite materials, often favorably distinguished by their performance, optical, chemical, mechanical, catalytic, electrical, and other characteristics. Therefore, detailed studies on synthesis and applications of NP-based materials is of great interest (Liu et al. 2017; Ahmad et al. 2015; Merino et al. 2015; Khalil et al. 2016; Rie and Thielemans 2017; Ferhan and Kim 2016).

Among NPs, gold (AuNPs) and silver nanoparticles (AgNPs) are quite interested and widely studied. First of all, this is due to their unique optical properties. They are based on the surface plasmon resonance phenomenon (SPR) (Amendola et al. 2017; Dykman et al. 2008; Dykman and Bogatyrev 2007; Wang and Ma, 2009; Ghosh and Pal 2007; Saarinen et al. 2009). It leads to appearance of the SPR bands in their absorption spectra. The optical properties of AuNPs and AgNPs are widely used in various fields of science and technology. In particular, these NPs are used to develop the spectrophotometric and colorimetric methods of analytical chemistry for the determination of various substances (Apyari et al. 2014; Terenteva et al. 2017; Priyadarshini and Pradhan 2017; Majdalawieh et al. 2014; Howes et al. 2014; Zhou et al. 2015). Size, shape and chemical surrounding of the NPs play a crucial role for their analytical properties and range of possible use. Therefore, a special attention should be paid to the method of synthesis and processes during NP formation.

Formation of nanoparticles proceeds through a series of successive stages: the emergence of individual atoms; nucleation and formation of an initial atomic cluster; growth of the cluster to a certain size; stabilization of NPs. Dimensions and dispersity of the NPs formed, as well as their stability over time, are regulated by varying the nature of the stabilizer and the synthetic process conditions. As stabilizers in the synthesis of monodisperse NPs, one can use excess of a reducing agent as well as specially introduced substances: ionic surfactants, e.g. sodium dodecyl sulfate or lauryltrimethylammonium chloride, ionic liquids, synthetic and natural polymers—polyvinylpyrrolidone, polyethylene glycol, cyclodextrins, chitosan and others (Dykman and Bogatyrev 2007).

As the reducing agent, sodium borohydride is used most often (Tan and Cheong 2013). Main feature of this widely used reductant is its high reduction ability, which makes it possible to obtain NPs of various metals differing greatly in their redox potentials. Sodium borohydride is quite stable in solid form and is convenient for use in the synthesis of nanoparticles in both aqueous and organic media.

The present work describes a method for preparation of polyurethane foam modified with borohydride as a new form of the widely known reducing agent and its possibilities for synthesis of metal nanoparticles.

Experimental

Reagents and instruments

The following reagents were used in the study: sodium borohydride, sodium hydroxide, cetyltrimethylammonium bromide (CTAB), hydrogen tetrachloroaurate, silver nitrate, hydrochloric acid, cysteamine hydrochloride. All they were at least of analytical grade. Stock solutions of these reagents were prepared using deionized water.

Polyether-based polyurethane foam (PUF) was cut into cylindrical tablets of 16 mm in diameter of (20 ± 2) mg each from an industrial sheet of the polymer. PUF tablets were cleaned in acetone by shaking for 10 min. The cleaning was repeated twice. After that the tablets were dried under stream of air. They were stored in a place protected from light.

Absorbances were recorded by SF-103 spectrophotometer (Akvilon, Russia), diffuse reflectance measurements were carried out using Eye-One Pro mini-spectrophotometer (X-Rite) (Apyari et al. 2011, 2013) on a white base.

The content of NPs in phase of polyurethane foam was estimated using values of the Kubelka–Munk function (F) at a wavelength of an absorption maximum (λmax): \(F = \frac{{(1 - R^{2} )}}{2R}\), here R is the diffuse reflection. pH was measured using Ekspert 001 pH-meter (Ekoniks Ekspert). TEM-images were recorded using a transmission electron microscope Libra 200 (Zeiss, Germany). The accelerating voltage was 200 kV. Samples were deposited onto copper grid support with a formvar film covered by amorphous carbon Formvar®/Carbon Reinforced Copper Grids 3440C-MB (SPI, USA). Scanning electron microscopic studies of PUF samples microstructure was carried out using a scanning electron microscope JSM 7100 F (Jeol, Japan) at the accelerating voltage of 2–5 kV.

Deionized water was obtained using the Millipore Simplicity purification system (Millipore). A mechanical shaker was also used.

Preparation of borohydride-modified polyurethane foam (PUF/BH4)

The polyurethane foam modified with borohydride was prepared by a sorption method. Sorption was carried out in a static mode. The PUF tablets were placed in a test-tube containing 5 mL of an aqueous solution of the following composition: 0.01 mol L−1 sodium borohydride, 0.01 mol L−1 sodium hydroxide and CTAB (0.0001 mol L−1 for the synthesis of nanocomposites or 0.01 mol L−1 for the synthesis of NPs colloids). The tablets were thoroughly pressed with a glass rod to remove air bubbles from pores and shaken on an electromechanical shaker for 15 min. After that, the tablets of PUF/BH4 were removed and dried between sheets of filter paper.

Synthesis of metal nanoparticles and nanocomposites with PUF/BH4 as a reducing agent

For the synthesis of metal NPs (gold and silver), freshly prepared PUF/BH4 tablets were placed in 5 mL of a suitable precursor solution (hydrogen tetrachloroaurate or silver nitrate respectively) of the concentration of 20 μg mL−1 in terms of atomic metal. The tablets were thoroughly pressed with a glass rod and stirred by shaking on an electromechanical shaker for 30 min. Then, the tablets were removed and dried between sheets of filter paper. Depending on the concentration of CTAB used during preparation of PUF/BH4, either NPs colloids or PUF/NPs nanocomposites can be prepared.

Analytical application of the PUF/AuNPs nanocomposites for the determination of cysteamine

To build the calibration curve, different amounts of cysteamine (0–26 μ mol L−1) were added during the synthesis of nanocomposites. After preparation of PUF/AuNPs, the diffuse reflectance of the resulting tablets was measured. The calibration curve was plotted as the decrease of the Kubelka–Munk function compared to the blank without cysteamine (ΔF) versus the concentration of cysteamine.

Results and discussion

General approach and scheme of the experiments

In this study, polyurethane foam was proposed as a solid matrix for preparation of a new form of a reducing agent for the synthesis of NPs. This choice was primarily justified by the unique sorption properties of this material, caused by the presence in its structure groups of different nature—urethane, ether, carbonyl, amino, phenylene, and alkyl. Owing to them, PUF effectively absorbs substances of different polarity, dissolving them in its polymeric membranes (Dmitrienko and Apyari 2010; Dmitrienko et al. 2012; Ramazanova et al. 2013). Additional advantages of this sorbent are the continuity and softness of the structure. These properties make it easy to separate the polymer from solution and remove residual liquid from pores by compressing between the sheets of filter paper. High efficiency of PUF sorption centers allows anchoring its network to not only chemical compounds, but also nanoparticles (Jain and Pradeep 2005; Jeon et al. 2008; Chou et al. 2006; Apyari et al. 2012; Gorbunova et al. 2017; Furletov et al. 2018; Arkhipova et al. 2015). This reveals possibility of creating nanocomposites based on this sorbent, which are promising for use in chemical analysis, medicine, healthcare, and other spheres. Moreover, polyurethane foam can be industrially produced as large soft and flexible pieces of various forms and reversibly compressed mechanically for many times. This opens prospects for application of the proposed approach for bulk preparation of the nanocomposites at an industrial scale.

In this work, PUF has been first modified with borohydride and studied as a new form of this widely known reductant for the synthesis of nanoparticles. The general scheme of the experiment is illustrated in Fig. 1a. It included two stages. In the first stage, sorption of sodium borohydride on PUF was carried out from a solution containing sodium hydroxide and CTAB. Sodium hydroxide is necessary as a stabilizer which reduces the concentration of hydrogen ions and, accordingly, the rate of borohydride decomposition. CTAB was added as an ion-pair reagent to form a hydrophobic ionic associate with borohydride anion to facilitate its transition to the polyurethane phase. The second step involved interaction of PUF/BH4 with hydrogen tetrachloroaurate (or other precursor), resulting in the formation of PUF nanocomposite with AuNPs (or AuNPs colloids depending on the conditions, as discussed below). In Fig. 1b the spectrum of the resulting nanocomposite of PUF with AuNPs (PUF/AuNPs) is shown. One can see a characteristic SPR band with a maximum at 540 nm in the spectrum of the nanocomposite. The shift of this band of about 20 nm in comparison with the characteristic SPR band of AuNPs in solution located at 520 nm is probably due to the influence of the polymer matrix on the plasmon oscillation frequency, as well as the more compact arrangement of AuNPs on the polymer surface compared to the aqueous solution.

Fig. 1
figure 1

a General scheme of the experiments; b diffuse reflectance spectrum of PUF/AuNPs nanocomposite

Full size image

The supposed mechanism includes slow diffusion-controlled desorption of borohydride anions from polyurethane foam and their interaction in the surface layer with a precursor to form NPs. In absence of a stabilizer that facilitates NPs transition to a colloidal solution, they attach onto the surface of the polymer, which in this case, in fact, plays a role of a solid-phase stabilizer, forming the nanocomposite.

PUF/BH4 synthesis optimization

The synthesis conditions have a significant effect on characteristics of the borohydride-modified polyurethane foam and its possibilities for the synthesis of nanoparticles.

First of all, modifying of PUF with borohydride can be accompanied with trapping of the solution drops containing redundant amounts of this reagent. This excess will affect reproducibility of PUF/BH4 properties. To choose appropriate initial conditions for preparation of PUF/BH4, a reproducibility study was done. Optical properties of the nanoparticles formed in the PUF phase were used to control their state. The data derived are represented in Table 1. One can see that high concentrations of the reagents (first three data rows) result in low reproducibility (the relative standard deviation, RSD of about 8–19%). This is assumed due to the redundant amounts of the reagents remaining on PUF that can be hardly eliminated by pressing of PUF samples with filter paper. Decreased concentrations of NaBH4 and CTAB (data rows 4–5) lead to improved reproducibility, however intensity of the spectrum becomes insufficiently low. Finally, a solution containing 0.01 mol L−1 NaBH4 and NaOH, and 0.0001 mol L−1 CTAB (data row 6) was considered as appropriate for preparation of PUF/BH4 with RSD of 6.6%.

Table 1 Reproducibility of PUF/AuNPs preparation at different concentrations of the reagents
Full size table

Effect of the phases contact time and the concentration of reagents on sorption of borohydride onto PUF was studied. Effect of these factors on the intensity of an SPR band of the resulting nanocomposite is illustrated in Fig. 2.

Fig. 2
figure 2

Dependence of F on phases contact time (a), concentration of NaBH4 (b), and CTAB (c). Meanings of F of the nanocomposite and absorbance of the solution at various concentrations of CTAB (d). Dependence of F of the nanocomposite on absorbance of the solution (e)

Full size image

From the data represented, it is evident that about 2/3 of the maximum possibly amount of sorbed borohydride is extracted within 1 min (Fig. 2a). This indicates a high rate of sorption, which is valuable from a practical point of view. To achieve the maximum signal, 10–15 min of the phase contact is required. In all subsequent experiments, sorption was carried out for 15 min.

An important factor affecting the characteristics of PUF/BH4 is the concentration of sodium borohydride in the solution for modification (Fig. 2b). With increasing concentration of NaBH4 up to 0.003 M, the number of NPs produced increases, which can be used to regulate optical and other properties of the resulting material. At high concentrations of NaBH4, the signal growth stops because of the full consumption of the precursor. Introducing large amounts of NaBH4 (> 0.01 M) is impractical because of both consumption of the reagent and the greater contribution of the surface-deposited reductant, amount of which is difficult to control, impairing reproducibility of the final material. To achieve the best reproducibility, NaBH4 concentration was chosen as 0.01 mol L−1 in subsequent experiments.

Unlike other factors, the concentration of alkali does not significantly affect the amount of gold nanoparticles formed. However, this reagent is necessary for stabilization of borohydride, which has a positive effect on reproducibility of the results. Therefore, in further experiments, its concentration was set to 0.01 mol L−1.

The role of CTAB is quite interesting. On the one hand, this substance promotes sorption of borohydride anions on the polyurethane foam, forming a hydrophobic ionic associate. On the other hand, it is known that CTAB is a stabilizer of nanoparticles (Tan and Cheong 2013) that promote formation of a colloidal system in solution at high concentration. It was shown (Fig. 2c) that increasing its concentration up to 0.001 M leads to an increase in the optical signal of the nanocomposite. At the same time, one can observe increasing formation of gold nanoparticles in the solution (Fig. 2d). At concentrations > 0.001 M, a decrease in the nanoparticles signal in both nanocomposite and solution is observed. However, a band of Au(III) complexes at 400–430 nm grows in the solution, i.e. formation of nanoparticles is suppressed under these conditions.

Increasing NaBH4 concentration up to 0.1 M leads to a sharper dependence of the nanocomposite absorbance on concentration of CTAB. Figure 2e illustrates that, depending on the concentration of CTAB, the higher amount of gold nanoparticles on PUF, the lower their amount in the solution is, and vice versa. At CTAB concentration of 0 and 0.1 M, gold nanoparticles preferably exist in the aqueous solution, at 0.0001–0.001 M CTAB they form the nanocomposite. In fact, one can assume that there is a competition between two stabilizers—a solid phase one (PUF) and a dissolved one (CTAB). This can be proved by an antibate linear relation between Kubelka–Munk function of the nanocomposite and absorbance of the solution (Fig. 2f). It shows the possibility of controlling distribution of NPs between phases.

On the basis of experimental data, the following conditions were chosen for the preparation of PUF/BH4 as a new form of a reducing agent for synthesis of nanocomposites: NaBH4 sorption time—15 min, NaBH4 concentration—0.01 M, sodium hydroxide and CTAB concentrations—0.01 and 0.0001 M, respectively. For synthesis of NPs colloids, the concentrations of NaBH4 and CTAB should be increased up to 0.1 M.

Stability of PUF/BH4

From a practical point of view, it is important to know stability of a reagent during storage. It is well known that aqueous solutions of NaBH4 are unstable and should be used immediately after preparation. It was of interest to study stability of PUF/BH4 in time (Fig. 3). It was shown that PUF modified in aqueous solution is stable only for several hours and loses its activity in a day (Fig. 3a). By analogy with the aqueous solution, it can be assumed that this is because of hydrolysis of borohydride by residual water in PUF structure. It was found that the modification from alcohol allows achieving increase in stability of the reagent. In this case, its activity remains at the same level even after 1 day (Fig. 3a).

Fig. 3
figure 3

The time stability of PUF/BH4 obtained in aqueous and ethanol solution when stored at room temperature (a); and PUF/BH4 obtained from an aqueous solution when stored in a freezer (b)

Full size image

The decomposition rate of borohydride can be reduced by cooling. From Fig. 3b it is clear that storage at low temperature (− 20 °C) in a freezer allows this reagent to be used for at least 1 month without any loss of activity (Fig. 3b), and even after half a year, the loss is not significant.

Thus, PUF/BH4 obtained even from aqueous solution is stable at low temperature and can be used for a long time. Preparation from aqueous solution is economically more justified, so synthesis PUF/BH4 from alcohol is unnecessary except for some special cases.

Application of PUF/BH4 for synthesis of metal nanocomposites

The new solid-phase reagent can be used to synthesize nanocomposites of metals. We have studied preparation of nanocomposites of gold and silver. Formed nanoparticles are attached on a polymer network. From a magnified image, by the example of a PUF/AuNPs cutoff shown in Fig. 4, it can be seen that the most intense coloration of polymeric membranes occurs at the edges. Within individual nodes, being cut perpendicularly to the direction of vision, the color is much weaker than around. This suggests that NPs are located on the surface of the polymer membranes and are practically absent inside. A SEM image represented in Fig. 4 shows that they are evenly distributed over the surface of the polymer. In our opinion, such a monolithic honeycomb structure of PUF/NPs is very interesting for practice. In particular, it may prove promising for the development of catalysts, sorbents, sensors, etc.

Fig. 4
figure 4

A sample of PUF/AuNPs nanocomposite (a), a magnified visible image of its structure (b) and its SEM image (c)

Full size image

Similarly, using an AgNO3 precursor, nanocomposites of PUF and silver NPs can be obtained. Electron microscopy showed that NPs on PUF surface are basically spherical in shape (Fig. 5). Their average diameters were 12 ± 5 and 34 ± 14 nm for gold and silver, respectively.

Fig. 5
figure 5

A SEM image of silver nanocomposite with PUF

Full size image

Figure 6 shows histograms of size distribution of NPs on PUF surface. Histograms can be characterized by a distribution close to normal. Polydispersity indices of these NPs calculated as PDI = s2/ <d> 2 (here s is the standard deviation of the diameter, <d> is the average diameter) are 0.0364 and 0.0413 for gold and silver, respectively.

Fig. 6
figure 6

Size distribution histograms of the gold (a) and silver (b) nanoparticles on PUF surface

Full size image

All obtained nanocomposites possess characteristic resonance optical properties inherent in the corresponding nanoparticles. Diffuse reflectance spectra of the nanocomposites obtained are given in Fig. 7. The maxima of the SPR bands of gold and silver are at 540 and 430 nm, respectively. The presence of optical properties in nanocomposites allows their possible use as sensitive materials of optical sensors.

Fig. 7
figure 7

Diffuse reflectance spectra of gold and silver nanocomposites with PUF

Full size image

Application of PUF/BH4 for synthesis of metal nanoparticles colloids

As was said above, control of distribution of the formed NPs between PUF and solution can be carried out by varying the concentration of CTAB and NaBH4 during preparation of PUF/BH4. Addition of excess CTAB and 0.1 M sodium borohydride leads to a colloid of nanoparticles (Fig. 1). No nanocomposite formation can be observed in this case. By the example of gold, characteristics of the NPs formed in solution were studied. Figure 8a shows a TEM image of AuNPs synthesized using PUF/BH4. In Fig. 8b, c TEM images of AuNPs obtained under the similar conditions but in the absence of PUF are given. In these cases, NaBH4 was added as an aqueous solution directly into the reaction mixture. It can be seen that the NPs obtained with the use of PUF/BH4 have a shape close to spherical and approximately equal size. Their average diameter is (4 ± 1) nm. The size distribution of these NPs is much narrower than when solutions of the corresponding reagents are mixed in the absence of PUF (Fig. 8d). The calculated polydispersity index (PDI) for AuNPs obtained using PUF/BH4 was 0.0264. This value is 5–10 times smaller than for the cases shown in Fig. 8b,c (PDI = 0.1069 and 0.2176, respectively). This value is also 10 times less than for AuNPs obtained by the classical Turkevich method (PDI = 0.282) (Manson et al. 2011; Kimling et al. 2006). Apparently, this is due to the slow release of borohydride during desorption from PUF, which avoids high local supersaturation and formation of many small particles. On the other hand, this can be ascribed to the stabilizing action of the polymer matrix with respect to nanoparticles forming near its surface. Thus, PUF/BH4 is promising for obtaining not only nanocomposites based on PUF, but also colloid solutions of NPs, which significantly expands its practical applications.

Fig. 8
figure 8

TEM images (a–c) and histograms of size distribution (d) of AuNPs in solution prepared using a PUF/BH4; b by adding solution of NaBH4 to solution of HAuCl4; c by adding solution of HAuCl4 to solution of NaBH4

Full size image

Practical application in analytical chemistry

Optical properties of PUF nanocomposites with metal NPs open a possibility of using PUF/BH4 as an analytical reagent. On the one hand, it can be used to determine gold and silver, on the other hand—substances that affect formation of NPs or their distribution between the polymer and the solution. An important advantage of this reagent is an ability to detect an analytical response using simple equipment or visually. At the same time, the stabilizing effect of PUF in relation to the formed NPs should provide good analytical performance and, accordingly, the ability to determine small contents of an analyte.

Within the second option, a possibility of determining cysteamine using PUF/BH4 has been demonstrated. Cysteamine is able to form a strong bond with gold due to the presence of the mercapto and the amino group. Addition of this compound at the stage of interaction between PUF/BH4 and HAuCl4 influences formation of the nanocomposite, which is manifested by a decrease in the SPR band and color of the sample (Fig. 9a). This effect can be used for cysteamine determination. Decrease of the Kubelka–Munk function at 540 nm was used as an analytical signal. I was shown that it depends linearly on the concentration of cysteamine in solution, which was used to construct a calibration curve (Fig. 9b). The method allows to determine the content of cysteamine in the range of 1.8–18 μmol L−1, the detection limit is 0.6 μmol L−1, which indicates good sensitivity of the determination.

Fig. 9
figure 9

Diffuse reflectance spectra and images of the nanocomposite samples obtained in the presence of different concentrations (μmol L−1) of cysteamine (a) and the calibration curve for the determination of cysteamine (b). \(\text{C}_{{\text{HAuCl}}_{4}}\) = 0.05 mM, t = 30 min, pH 3.7, b λ = 540 nm

Full size image

Conclusions

Borohydride-modified polyurethane foam has been proposed as a new form of the well-known reducing agent to obtain nanocomposites and colloidal solutions of nanoparticles of various metals. This solid-phase reagent can be obtained by a simple sorption technique. Due to the controlled desorption of borohydride from polyurethane foam, the synthesized nanoparticles has a narrow size distribution. Control of nanoparticles distribution between polymer and solution can be easily performed by varying CTAB and NaBH4 concentration during preparation of the borohydride-modified polyurethane foam. The change in distribution of nanoparticles between the phases in the presence of cysteamine makes it possible to propose an approach to its determination using diffuse reflectance spectroscopy. The advantages of the proposed scheme include simplicity, high throughput, availability of the equipment used, and facility of a semi-quantitative test-version implementation. Synthesized nanocomposites have a potential to be used in medicine, catalysis, analytical chemistry.