Fabrication of silver nanoparticles through reduction by ε-caprolactam without using protecting agent

Abstract

ε-Caprolactam (CL) was used as a multifunctional medium to synthesize silver nanoparticles (Ag NPs). In this method, CL played three key roles including reducing agent, protecting agent and solvent. The identity of the solid substance generated in molten CL, silver nanoparticles (Ag NPs), was verified by X-ray diffraction (XRD). The particle size of the as-synthesized Ag NPs was uniform and less than 15 nm. Coordination of CL with silver ions/atoms and moderate reduction contributed together to the resultant nanosized silver particles.

---

Applications of the significant features of silver have been of continuing interest since ancient times. Because of the controllable size distributions and shapes, as well as a large surface area,¹ silver nanoparticles (Ag NPs) have potential applications such as in novel electrical, optical, magnetic, catalytic, and sensing technologies.²

Since Faraday first presented the preparation of metal nanoparticles in aqueous medium,³ a large number of methods have been developed for the synthesis of Ag NPs. Besides the irradiation of the solution containing silver ions with ultraviolet, visible light or microwave,⁴ the most important strategy for Ag NPs fabrication is the chemical reduction of precursors by reducing agent such as sodium borohydride,⁵ citrate,⁶ polyalols,⁷ N,N-dimethylformamide,⁸ as well as polysaccharide.⁹ In order to fabricate Ag NPs with ideal size and shape in aqueous medium, the reducing agent, protecting agent and solvent are the most important factors and should be carefully chosen.

ε-Caprolactam (CL) has high polarity, high boiling point, good coordinating capacity, and is relatively stable at moderate temperature. We have demonstrated previously that it can act as an effective medium for NPs dispersion and guarantee uniform distribution of NPs in the resultant in situ polymerized product of polyamide 6 matrix.¹⁰ When we precipitated metallic oxide NPs from molten CL for the first time,¹¹ it seemed that CL was likely to be a good medium for many metal salts,¹² which would facilitate the fabrication of NPs in molten CL.

In this paper, a brand new route for Ag NPs fabrication is presented in which neither extra solvent nor extra protecting agent was needed.¹³ The CL itself played all three key roles: solvent, protecting agent and reducing agent, a process that has rarely been reported previously.

In a typical synthesis, different dosages of AgNO₃ (0.085–0.255 g) were added to 20 g CL melt at 140 °C. The mixture was stirred for 16 h under the protection of argon (Ar). The precipitate was collected by alternately wash with ethanol and centrifugation for three rounds. The suspension containing Ag NPs was characterized by TEM, and the results are illustrated in Fig. 1. The size of Ag NPs was less than 15 nm for all three samples. It can be seen that the weight ratio between CL and AgNO₃ strongly influenced the size of the resultant particles. All of the particles were less than 7.5 nm when the dosage of AgNO₃ was 0.085 g. About 85% particles were spread between 2.5 and 6 nm. When the dosage of AgNO₃ was increased to 0.17 g, the size distribution became narrower and the size of most particles was in the range of 8–13 nm. The percentage of size between 8 and 12.5 nm was over 85%. The size distribution was relatively narrow with a standard deviation of 9%. With a further increase of dosage of AgNO₃, some particles with size less than 7.5 nm were generated. Compared with the former sample, although the particle size spanned the same range, the distribution was altered, and the standard deviation was 28%.

Fig. 1 TEM micrographs and particle size distribution of resultant Ag NPs. TEM graph from (a) to (c) correspond to Ag NPs prepared with different dosage of AgNO₃: (a) 0.085 g, (b) 0.17 g and (c) 0.255 g.

Precious metals are known for the possibility of sintering at relatively lower temperature when synthesized in organic or inorganic media.¹⁴ As we performed at relatively high temperatures (140 °C), the formation of the particles could be a result of particle–particle sintering during reaction. Close observation of the TEM in Fig. 1 showed no sintered particles. The X-ray diffraction pattern of Ag NPs prepared at the dosage of AgNO₃ 0.17 g is shown in Fig. 2. The peaks centered at 2-theta corresponded to 38°, 44°, 64°, 77° and 81° corresponding to (111), (200), (220), (311), and (222) planes respectively.¹⁵ The powder exhibits good crystallinity. According to the mathematical deconvolution of the peaks, the particle size can be calculated according to Scherrer formula.¹⁴ Width at half maximum intensity of (111) plane indicated a crystallite size of 8 nm. This approximate value was close to the mean particle sizes calculated by image analyses of our colloids (ca. 11 nm). Since Scherrer formula always tends to underestimate the real crystallite size,¹⁴ this result seems to favor the hypothesis of the monocrystallinity of the particles.

Fig. 2 XRD pattern of the synthesized Ag NPs.

When the temperature was elevated to 140 °C, the appearance of the mixture of AgNO₃ and CL changed gradually from clear and transparent at beginning, to yellow, and then to opaque black at the end, suggesting thermal reduction of AgNO₃ in molten CL. Previous reports of the change of graphene oxide to graphene by decomposition of labile oxygen-contained moieties in the molten CL after the treatment at 250 °C for 8 hours was the indirect evidence for the reducibility of CL.¹⁶ The shiny silver mirror on the inner surface of flask at the end of the reaction also indicated the formation of metallic silver.

The upper transparent solution was characterized by UV-vis spectra as shown in Fig. 3. For comparison, UV-vis spectra of pure CL treated at different conditions were also recorded. Pure CL treated at 140 °C for 16 hours under the protection of Ar showed no absorption at range of 220–500 nm (curve 1). However, when pure CL was treated by oxygen (O₂, 30 ml min⁻¹) at 140 °C for 16 hours without addition of AgNO₃, an intensive absorption at 240 nm appeared (curve 3). This means that at 140 °C, O₂ could oxidize CL, and the new absorption peak at 240 nm corresponds to the oxidation product of CL. The same absorption peak at 240 nm was also observed when AgNO₃ was added to molten CL after the reaction sustained at 140 °C for 16 h under protection of Ar (curve 2). The absence of absorption at 240 nm in the case of pure CL protected by Ar indicates that CL experienced no chemical conversion (curve 1). Similar absorption at 240 nm for CL treated by different oxidizing agents (O₂ and AgNO₃, curve 3 and curve 2) illustrates that the same oxidation product was generated, and as a result, we can conclude that reduction of AgNO₃ can be initiated by CL at 140 °C. Furthermore, the intensity of the absorption peak of oxidized CL was 0.75 when treated by AgNO₃ and 0.87 when treated by O₂, the sum of which was almost equal to the absorption intensity of 1.54 of the oxidized CL treated by AgNO₃ and O₂ together (curve 4). Thus, it can be deduced that in the preparation of Ag NPs, CL played the role of reducing agent and initiated the reduction of AgNO₃ at 140 °C.

Fig. 3 UV-vis spectra of CL treated at different conditions. Curve 1 represents the pure CL treated under the protection of Ar. Curve 2 and curve 3 illustrate the CL treated by AgNO₃ and by O₂, respectively. Curve 4 corresponds to the CL treated by AgNO₃ and O₂ together.

In previous studies of the thermal oxidation of CL initiated by O₂, it was found that products of this reaction were caprolactam hydroperoxide, adipimide and adipic acid monoamide.¹⁷ The kinetics analysis of thermally initiated oxidation of CL indicated that caprolactam hydroperoxide was the primary product, and adipimide and adipic acid arose from consecutive decomposition reactions of caprolactam hydroperoxide.¹⁸ The UV-spectra in Fig. 3 indicate that CL may be oxidized by AgNO₃ in a mechanism similar to that of the oxidation of CL initiated by O₂. Because the carboxylic group of adipic acid monoamide was the inhibitor of the oxidation of caprolactam,¹⁸ only low yields of oxidation products of CL were reported.¹⁹ Although there are several published works analyzing the impurities and oxidation products of CL,²⁰ the results are confusing. ¹³C NMR spectra of pure CL, CL oxidized by AgNO₃ and O₂ showed no difference in the chemical shift in Fig. 1S.† Maybe this explains why the studies of oxidation of CL provided little statistical information about oxidation products of CL.^19,20

CL is the precursor monomer of polyamide 6, which is a well-known condensation-type polymer with extensive applications in the field of fibers and engineering materials.²¹ In the in situ preparation of NP-filled polyamide composites, CL was always used as dispersant of NPs without using other solvents. As a result, potential particle aggregation of NPs caused by solvent evaporation was effectively avoided. Many metal salts could be completely dissolved in molten CL, and metallic oxide NPs could be generated when a precipitator (sodium hydroxide) was added.¹¹

In addition, the test of conductivity showed that the mixture of AgNO₃ and molten CL was ∼100 μS cm⁻¹, while the conductivity of silver nitrate aqueous solution was greater than 10⁵ μS cm⁻¹ at the same concentration. This indicated that AgNO₃ was partly ionized in molten CL, but the concentration of silver ions from AgNO₃ in molten CL was very low. Unionized AgNO₃ was the stock of silver ions, which continuously replenished the silver ions after they were consumed through the reduction by CL. The little variation in conductivity of the mixture after 120 min indicated that the concentration of silver ions did not vary much. As a consequence, the low and stable concentration of silver ions kept the reaction of AgNO₃ with CL at a moderate rate. Combining the coordination between silver ions/atoms and CL and the moderated rate of reduction of AgNO₃, leads to production of nanosize silver particles.

Conclusions

In summary, we have presented a novel method employing CL as not only solvent, but also reducing agent and stabilizer for synthesis of Ag NPs. What's more, Ag NPs with homogeneous size were guaranteed (as is shown by TEM) because of the coordination effect of CL with silver ion/atom and moderate rate of reduction of AgNO₃.

Acknowledgements

This work is financially supported by the Shanghai Genius Advanced Materials Co. Ltd.

References

1. (a) B. J. Wiley, D. Ginger and Y. Xia, Nano Lett., 2007, 7, 1032; (b) B. J. Wiley, D. H. Cobden and Y. Xia, Nano Lett., 2006, 6, 2273; (c) Y. Xia, P. Yang and Y. Sun, Adv. Mater., 2003, 15, 353; (d) Y. Sun, T. Herricks and Y. Xia, Nano Lett., 2003, 3, 955.
2. (a) X. Xia, J. Zeng and Y. Xia, J. Phys. Chem. C, 2012, 116, 21647; (b) J. Zeng, S. Roberts and Y. Xia, Chem.–Eur. J., 2010, 16, 12559; (c) N. C. Martins, C. S. Freire and T. Trindade, Cellulose, 2012, 19, 1425; (d) R. M. El-Shishtawy and A. M. Asiri, Cellulose, 2011, 18, 75.
3. M. Faraday, Philos. Trans. R. Soc. London, 1857, 147, 145.
4. (a) Y. Sun and Y. Xia, Adv. Mater., 2003, 15, 695; (b) R. Jin, Y. Cao and J. Zheng, Science, 2001, 294, 1901; (c) S. Kamel, Carbohydr. Polym., 2012, 90, 1538.
5. T. Huang and H. Xu, J. Mater. Chem., 2010, 20, 9867.
6. J. Zeng, J. Tao and Y. Xia, Chem.–Asian J., 2011, 6, 376.
7. S. E. Skrabalak, B. J. Wiley and Y. Xia, Nano Lett., 2008, 8, 2077.
8. (a) I. Pastoriza-Santos and L. M. Liz-Marzán, Nano Lett., 2002, 2, 903; (b) I. Pastoriza-Santos and L. M. Liz-Marzán, Langmuir, 1999, 15, 948.
9. (a) D. Wei and W. Qian, Colloids Surf., B, 2008, 62, 136; (b) B. Xia, Q. Cui and L. Li, Langmuir, 2012, 28, 11188.
10. (a) Y. Liu, Z. Chen and G. Yang, J. Mater. Sci., 2011, 46, 882; (b) Y. Liu, T. Xie and G. Yang, J. Mater. Sci., 2011, 46, 5050; (c) Y. Liu and G. Yang, Thermochim. Acta, 2010, 500, 13.
11. H. Xia and G. Yang, J. Mater. Chem., 2012, 22, 18664.
12. (a) H. Xia, X. Zhao and G. Yang, Mater. Lett., 2013, 98, 90; (b) H. Xia and G. Yang, RSC Adv., 2013, 3, 12320.
13. (a) R. Jin, G. C. Schatz and C. A. Mirkin, Nature, 2003, 425, 487; (b) M. Oh and C. A. Mirkin, Nature, 2005, 438, 651; (c) Y. Sun and Y. Xia, Nature, 1991, 353, 737; (d) X. Wang, J. Zhuang and Y. Li, Nature, 2005, 437, 121.
14. (a) P. Scherrer, Mathematisch-physikalische Klasse, 1918, 1918, 98; (b) P. Y. Silvert, R. Herrera Urbina and K. Tekaia-Elhsissen, J. Mater. Chem., 1997, 7, 293.
15. P.-Y. Silvert, R. Herrera-Urbina and K. T. Elhsissen, J. Mater. Chem., 1996, 6, 573.
16. Z. Xu and C. Gao, Macromolecules, 2010, 43, 6716.
17. A. Rieche and W. Schon, Chem. Ber., 1966, 99, 3238.
18. B. Lánská and J. Šebenda, Angew. Makromol. Chem., 1988, 164, 181–198.
19. G. Dege and H. Reimschuessel, J. Polym. Sci., Polym. Chem. Ed., 1973, 11, 873–896.
20. (a) B. Pant, H. R. Pant and H. Y. Kim, Colloids Surf., A, 2012, 395, 99; (b) K. Hashimoto, Prog. Polym. Sci., 2000, 25, 1411; (c) X. Zhang, X. Fan and H. Li, J. Mater. Chem., 2012, 22, 24081.
21. (a) M. C. Rycenga, M. Claire and Y. Xia, Chem. Rev., 2011, 111, 3669; (b) Q. Zhang, W. Li and L.-P. Wen, Chem.–Eur. J., 2010, 132, 11372.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra01675k

---