Journal of Nanoparticle Research

, 14:1185

A novel method for synthesis of 56Co-radiolabelled silica nanoparticles

Authors

  • I. Cydzik
    • European Commission, Joint Research CentreInstitute for Health and Consumer Protection
    • Institute of Nuclear Chemistry and Technology
    • Heavy Ion LaboratoryUniversity of Warsaw
  • A. Bilewicz
    • Institute of Nuclear Chemistry and Technology
  • K. Abbas
    • European Commission, Joint Research CentreInstitute for Transuranium Elements (Ispra Site)
  • F. Simonelli
    • European Commission, Joint Research CentreInstitute for Health and Consumer Protection
  • A. Bulgheroni
    • European Commission, Joint Research CentreInstitute for Health and Consumer Protection
    • European Commission, Joint Research CentreInstitute for Health and Consumer Protection
  • N. Gibson
    • European Commission, Joint Research CentreInstitute for Health and Consumer Protection
Research Paper

DOI: 10.1007/s11051-012-1185-x

Cite this article as:
Cydzik, I., Bilewicz, A., Abbas, K. et al. J Nanopart Res (2012) 14: 1185. doi:10.1007/s11051-012-1185-x

Abstract

A method for synthesis of radiolabelled amorphous silica nanoparticles is presented. The method is based on the well-known Stöber process with the exception that 56Co radiotracer is introduced into one of the precursor materials prior to the initiation of the nanoparticle synthesis. The 56Co was prepared by proton irradiation of an iron foil, followed by dissolution in hydrochloric acid and 56Co/Fe radiochemical separation. In order to determine the residual Fe in the 56Co radiotracer solution, ICP-MS measurements were performed. Nanoparticles in the size range 20–100 nm were synthesised and characterised by gamma spectrometry, ICP-MS, XRD, DLS, and Zeta potential measurement. It was shown that the size and Zeta potential of the nanoparticles was roughly the same following synthesis with or without added 56Co, and in both cases, the structure was that of amorphous silica. It was found that 99.5 % of the 56Co was bound into the nanoparticles during synthesis, and centrifugation experiments confirmed that the radiolabels were stably incorporated into the silica matrix.

Keywords

Silica nanoparticlesRadiolabelling of nanoparticlesRadiochemical synthesisRadiotracer

Introduction

The rapid increase in the use of nanoparticles (NPs) in industrial processes and their incorporation into consumer products has led in recent years to growing concern about possible risks for human health and the environment (Weiss and Diabate 2011; OECD 2008; Ju-Nam and Lead 2008; Handy et al. 2008; Hoet et al. 2004). Products and processes involving or containing nanoparticles will however continue to be developed, since the benefits are very promising in a wide range of applications (e.g. Schaefer 2010; Vollath 2008). It is important therefore to investigate whether significant risks actually exist regarding the production and use of different types of nanoparticles, and to develop appropriate methods for detection, characterisation and safety assessment (Weiss and Diabate 2011; Gibson et al. 2011; Oberdörster et al. 2005). Such investigations are of basic importance to support the safe and sustainable development of nanotechnology-based products.

Silica NPs have been used in a wide variety of applications for many decades including cosmetics, fabrication of electric and thermal insulators, food additives and drug delivery systems (e.g. Tang et al. 2012; Napierska et al. 2010; Grobe et al. 2008). The most commonly applied industrial process for their synthesis is based on flame pyrolysis of silicon tetrachloride or vaporisation of quartz sand in a 3,000 °C electric arc. The resulting powder is called ‘fumed silica’, consisting of aggregated/agglomerated primary particles of (typically) some tens of nm diameter (Barthel et al. 2008). The annual industrial production volume of silica NPs is estimated to be nowadays well above 1,100 kt (Lázaro and Brouwers (2010) data for 1999, Kammler et al. (2001)). Various other synthesis methods can be used to produce silica NPs (Jafarzadeh et al. 2009; Suzuki et al. 2004), some of which are suitable for small-scale laboratory synthesis, producing NP suspensions with a narrow size distribution. JRC IRMM Geel uses colloidal silica slurry as the starting material (Ludox TM50 supplied by Grace Davison (DE)) to produce IRMM-304 silica reference material (Couteau and Roebben 2008; Couteau et al. 2010). IRMM-304 is a quality control material (QCM) and consists of silica nanoparticles suspended in an aqueous solution with an average size of 40 nm.

For tracing studies, fluorescent labels can be incorporated into silica NPs (Mader et al. 2008; Wang et al. 2006), but fluorescent tracing has limitations with respect to sample preparation for subsequent optical detection, especially in systems such as complex organisms or in simulated environmental models. It also requires the incorporation of fluorescent molecules into the silica NPs (implying significant modification to the synthesis process), or their attachment to the surface (with possible modification to the NP properties). Radiolabelling of NPs has several advantages over fluorescent labelling, in particular with respect to simple sample preparation for detection and very high sensitivity (Gibson et al. 2011). However, silica cannot be effectively radiolabelled for tracing over several days or weeks by direct irradiation methods such as neutron or light-ion irradiation, since there is no nuclear reaction available with a high enough yield that leads to an isotope with a suitable half-life (see Table 1). With the exception of 32Si, there is also no intrinsic (Si or O) radioisotope suitable for SiO2 NP synthesis from radioactive precursors. However, 32Si is a pure β-emitter (no gamma emissions) with a very long half-life (172 years), so is less simple to trace, and presents more of a problem in terms of waste management due to the long half-life. It is possible to radiolabel silica by chelation and surface attachment of isotopes such as 99mTc, 177Lu, 188Re or 111In, but the applicability of such techniques is questionable in some cases, since in some environments the radiolabels might detach from the NPs, and in addition the labelling entails surface modification of the NPs which might influence their behaviour.

Short-term PET tracing studies of porous silica particles prepared by surface reactions with 18F have been reported (Sarparanta et al. 2011), while proton irradiation of silica nanoparticles enriched in 18O is also being investigated as a method for 18F labelling (Llop et al. 2011). Activation of larger oxide particles by 3He irradiation is an effective method for 18F labelling and single particle tracing for some industrial processes (Leadbeater et al. 2012). Irradiation by 3He of silica NPs, however, presents a number of prohibitive experimental difficulties, due to the high rate of energy loss of 3He both in target capsule windows and in the powder itself during irradiation (Gibson et al. 2011). For safety studies, it may be necessary to be able to trace silica NPs over extended time periods, especially if in vivo bio-distribution and fate should be determined. This makes 18F, with its short half-life, unsuitable as a radiotracer. For such studies, ideally it should be possible to determine NP concentrations in different organs over several weeks and to very low levels, for example down to 10−6 or less of the applied dose. For in vitro cell uptake studies, shorter time scales, of a few days, are relevant, but 18F is still not a suitable tracing isotope, while fluorescent labelling is not as powerful for quantitative assessment as radiolabelling. We undertook therefore experiments on the synthesis of silica NPs with 56Co as an incorporated extrinsic radiotracer isotope. 56Co was chosen, since it can be produced easily by proton irradiation of an iron target at a cyclotron facility (even using the mini-cyclotrons widely available throughout the World), it can be easily separated from the target material, and its half-life of 77.26 days and intense gamma emissions, make it very suitable for tracing experiments.
Table 1

Nuclear properties of selected Si and O isotopes

NP material

Radioisotope

Nuclear reaction

Half-life

Decay-mode

Comments

SiO2

26Si

28Si (p, t)

2.21 s

β+ to 26Al

Half-life too short

24Mg (3He, n)

24Mg(3He, n)

27Si

27Al (p, n)

4.14 s

β+ to 27Al

Half-life too short

31Si

30Si (n, γ)

2.62 h

β to 31P

Half-life too short for most studies, abundance of 30Si too low

30Si (d, p)

32Si

30Si (t, p)

172 years

β to 32P

Half-life too long, no γ rays

15O

14N (d, n) 15O

122 s

β+ to 15N

Half-life too short

15N (p, n) 15O

16O (p, pn) 15O

The comments refer to the radioisotopes suitability for safety/environmental tracing studies

Materials and experimental methods

Hydrochloric acid (HCl) 37 % was purchased from Fluka (Italy). Tetraethyl orthosilicate (TEOS) reagent grade 98 %, ammonium hydroxide 25 %, absolute ethanol ≥99.5 %, and acetone ≥99.8 % were purchased from Sigma-Aldrich (Italy). Phosphate-Buffered Saline 1X (PBS), pH 7.2 was obtained from Gibco Life Technologies (Italy). All chemicals were analytical grade and were used without further purification.

56Co radiotracer production

Proton irradiation of natural iron was used to produce 56Co at the Scanditronix MC40 cyclotron at JRC Ispra on a beam line equipped with a pneumatic remote sample transfer system between the beam line and the sample manipulation glove box. Irradiation with proton energy of below 20 MeV produces four radioisotopes of cobalt: 55Co (T1/2 = 17.54 h), 56Co (T1/2 = 77.26 days), 57Co (T1/2 = 271.79 days) and 58gCo (T1/2 = 70.86 days). The activation parameters were optimised for the production of 56Co which has a well-resolved intense gamma-ray peak at an energy of 846.7 keV with a high branching ratio (99.9 %) which allows very accurate activity measurements for this radioisotope. Furthermore, 56Co is produced in significant quantities by the (p,n) nuclear reaction in view of the fact that 56Fe is the most abundant (91.8 %) stable isotope in natural iron (de Laeter et al. 2003).

For the irradiation, an aluminium shuttle capsule as shown in Fig. 1 with an inner diameter of 10 mm was loaded with two iron foils (99. 99+ % pure) purchased from GoodFellow Cambridge, each of thickness 25 μm, diameter 10 mm, and mass 15 ± 0.15 mg. The impurity traces of the iron foil as indicated in the certificate of analysis were: Ag 1 ppm, Al 2 ppm, Ca 3 ppm, Cr 1 ppm, Cu 2 ppm, Mg 2 ppm, Mn 1 ppm, Ni 1 ppm, and Si 3 ppm. The energy of the proton beam was set to 16.5 MeV which results in a proton energy of 12.6 MeV on the iron target material after passing of the protons through a 300 μm beamline aluminium window and the 275 μm aluminium entrance window of the shuttle capsule. These beam energies have been calculated using the stopping and range of ions in matter (SRIM) code (Ziegler et al. 2008). Refrigerated helium cooling from the front (beam) side and water cooling from the back side of the capsule ensured that the material target did not overheat during irradiation. The iron foils were irradiated for 4 h with a beam current of 5 μA, to obtain a total activity of around 3,000 kBq of 56Co at the end of bombardment (EoB), after which the target was automatically sent to the transfer station located in the glove box. Following irradiation and a cooling down period of one day to allow shorter-lived isotopes to decay and reduce radiation exposure during sample manipulation, the iron foils were removed from the aluminium shuttle capsule and analysed with gamma spectrometry to estimate the 56Co activation yield and the presence of other isotopes.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-012-1185-x/MediaObjects/11051_2012_1185_Fig1_HTML.gif
Fig. 1

The aluminium shuttle capsule for iron foils irradiation: (1) water cooling from the back side of the capsule, (2) refrigerated helium cooling from the front (beam) side, (3) O-ring seal

56Co/Fe radiochemical separation

The obtained 56Co was separated from the Fe target material by an anion exchange process adapted from that reported by Hazan and Korkisch (1965). A Sigma-Aldrich glass column (9 × 0.5 cm) containing the strongly basic anion exchange resin Dowex 1X8 (Cl-form, particle size 0.075–0.15 mm, mesh size 100–200, suspended in deionised water) purchased from BDH Chemicals Ltd. Poole (England), was equilibrated with a mixture containing 90 % acetone and 10 % of 9 M HCl solution. Before use, the resin was washed again with approximately 100 mL of this solution. Then 30 mg of the activated Fe target material was dissolved in 1 mL of 9 M HCl and diluted with acetone to 10 mL. Afterwards, the dissolved target solution was introduced onto the ion exchange resin. The 56Co solution was passed through the column with a flow rate 0.2 mL/min. After adsorption of 56Co radiotracer, the anion exchanger column was washed with 20 mL acetone–6 M HCl mixture to remove all remaining Fe as Fe(II) ions. Then the 56Co radionuclide was eluted with 10 mL of 90 % acetone and 10 % of 3 M HCl solution in 1 mL fractions (Fig. 2). The Fe(III) ions are strongly bound to the exchanger resin at this HCl concentration and can be eluted only with 0.5 M HCl. The activity measurements in the successive stages of radiochemical separation were performed with gamma spectrometry.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-012-1185-x/MediaObjects/11051_2012_1185_Fig2_HTML.gif
Fig. 2

Schematic view of 56Co/Fe radiochemical separation procedure

The 56Co/Fe radiochemical separation on the Dowex 1X8 (Cl) resin proved to be very efficient. The 56Co activity loss during the passage of the dissolved Fe target through the column was less than 2 %. More than 93 % of the total 56Co activity was eluted in three initial 1 mL fractions (Fig. 3). The impurities of Fe2+ ions in the three initial 56Co radiotracer solution fractions were determined using the Inductively Coupled Plasma Mass Spectrometry technique (ICP–MS Sciex Elan 6100 DRC II, PerkinElmer, Canada).
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-012-1185-x/MediaObjects/11051_2012_1185_Fig3_HTML.gif
Fig. 3

Elution curve of 56Co with mixture of acetone–3 M HCl from ion-exchange column filled with Dowex 1X8 (Cl) resin

Radiochemical synthesis of 56Co-labelled SiO2 nanoparticles

Basing the process on the Stöber concept (Stober et al. 1968; Ibrahim et al. 2010), amorphous SiO2 NPs were synthesised in aqueous medium containing ammonia as catalyst. First, water/ammonia mixture at the appropriate concentration ratio (see Table 2) was placed in a glass flask and an appropriate quantity of absolute ethanol progressively added. The solution was stabilised at room temperature with slow stirring for 15 min to ensure complete mixing. Then, the required volume of tetraethyl orthosilicate (TEOS) in absolute ethanol (previously mixed) was carefully added. The reaction then took place under slow stirring for a time period of 5–24 h depending on the chosen synthesis conditions. During this chemical process, amorphous SiO2 NPs were formed by gradual diffusion of TEOS from the organic phase into the aqueous phase, where hydrolysis and condensation reactions took place. The obtained colloidal solution was separated by high-speed centrifugation, and the SiO2 NPs were washed once with ethanol and after that three times with deionised water. Variation of synthesis conditions, such as reaction time, TEOS and catalyst concentrations allows the production of monodisperse amorphous SiO2 NPs with stable size distribution in the range 20–100 nm (Table 2).
Table 2

Amorphous SiO2 NPs: synthesis conditions and obtained nominal concentrations

SiO2 NPs (nm)

Synthesis conditions

Nominal concentration (mg/mL)

20

TEOS: 0.05 M, NH3: 0.1 M, H2O: 1 M

1.43

40

TEOS: 0.2 M, NH3: 0.2 M, H2O: 1 M

1.78

100

TEOS: 0.2 M, NH3: 0.3 M, H2O: 1 M

3.50

Based on the same method, radioactive 56Co-labelled SiO2 NPs were synthesised. The initial volume of the 56Co solution was reduced (ca. 0.2 mL) by evaporation of the residuals of the eluent used during radiochemical separation procedure. Afterwards, the 56Co radiotracer was combined with 1 mL of deionised water and a suitable quantity of ammonium hydroxide to adjust the final pH of the solution to the range 6.8–7.2. In order to synthesise 56Co-radiolabelled silica nanoparticles, an appropriate activity concentration of obtained 56Co radiotracer was added to the water/ammonia phase of the reaction mixture. The subsequent steps of the 56Co–SiO2 NPs radiosynthesis were followed according to the synthesis method for non-active SiO2 NPs.

Nanoparticle characterisation

The 56Co radioisotope yield was determined by gamma spectrometry. High purity germanium (HPGe) detectors from CANBERRA (USA) and EG&G Ortec (USA) were used, calibrated with standard calibration sources in energy and efficiency for each sample geometry. The certified calibration standard sources were supplied by ENEA (Italy), Czech Metrological Institute, DAMRI, and CERCA (France). In order to obtain high accuracy, different geometries were used to minimise uncertainties due to dead time of the electronic signal processing chain of the gamma-ray acquisition system, depending on the activity of the specimens and the resulting count rate. The acquisition time was chosen to obtain sufficient counting statistics (below 1 % of uncertainty on the net peak area).

The size distribution of the non-active SiO2 and radiolabelled 56Co–SiO2 NPs in two different types of media was determined by dynamic light scattering (DLS) using a Zetasizer Nano ZS system (Malvern Instruments: Malvern, UK). The DLS technique also determines a ‘polydispersity index’ (PdI), which is an indicator of the degree of size variation of the particles tested. In order to verify the influence of dispersion medium on the aggregation/agglomeration state, SiO2 and 56Co–SiO2 NPs were suspended in Milli-Q water and Phosphate-Buffered Saline (PBS) solution. For DLS analysis, samples of 1.5 mL were obtained from 100 μg/mL stock solutions. After 180 s equilibration at 25 °C, each sample was measured three times, and the average value was calculated using the instrumental software. The duration for each measurement was set on ‘automatic’. All achieved DLS analysis results were based on intensity distribution data. In order to determine the surface charge of the silica NPs in suspension, Zeta potential measurements were performed on both the SiO2 and 56Co–SiO2 NPs samples, also using the 100 μg/mL stock solutions and the Malvern Zetasizer Nano ZS system. The structure of the non-active SiO2 and radiolabelled 56Co–SiO2 NPs was assessed using X-ray diffraction. The measurements were made on a dedicated glancing angle diffractometer by mixing a small amount of NP material with poly (methyl methacrylate) (PMMA) dissolved with anisole and smearing this onto the surface of a silicon wafer over an area of about 0.5 cm2. Owing to the high surface sensitivity of the glancing angle XRD technique, the amount of material analysed could be maintained low enough so that no special precautions regarding sample radioactivity had to be taken during the XRD measurements. Measurements were made with Cu Kα radiation and an instrumental resolution of ~0.2°.

Stability studies of the 56Co radiolabel within the SiO2 NPs were performed in both MilliQ-water and PBS solution for each NP size. Two millilitres of the appropriate medium were put into a centrifuge tube. A known amount of 56Co–SiO2 NPs activity in 0.5 mL was added. Then, the suspension was shaken for 30 s and agitated using an ultrasonic bath (Branson 2200, USA) for 5 min at room temperature. Taking into account the differences in the sedimentation rate between smaller and larger nanoparticles, 100 nm silica suspensions were centrifuged for 20 min at 4,000 rpm, whereas 40 and 20 nm suspensions for 30 min at 6,000 rpm at room temperature to obtain two samples—the supernatant and the separated pellet. The centrifuge used was an EBA 20 HETTICH Zentrifugen (Germany). The samples were removed for later gamma-spectrometry measurements. Volume corrections were done and the activity percentage was determined in each phase. The stability studies were repeated each day for 14 days at room temperature.

Results and discussion

Figure 4 shows the gamma spectrum of an irradiated iron foil. In addition to 56Co, the presence of 55Co and 57Co radioisotopes can also be seen in the gamma spectrum, but in very low quantities (0.3 %) with respect to the 56Co. The 511 keV gamma emission arises from the electron–positron annihilation, and is present in the gamma spectrum of all positron emitting radioisotopes. The achieved quantity of 56Co radiotracer was sufficient to obtain 56Co-radiolabelled SiO2 NPs with the level required for both in vitro cell uptake studies, or for in vivo biodistribution studies, or for other tracing experiments such as in simulated environmental systems. If higher activities were required, the irradiation current and/or time can easily be increased. The radioactivity level achieved following synthesis of the different sized NPs was 1.94 MBq/mg for the smallest (20 nm) NPs, 1.56 MBq/mg for the 40 nm NPs, and 1.28 MBq/mg for the largest (100 nm) NPs.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-012-1185-x/MediaObjects/11051_2012_1185_Fig4_HTML.gif
Fig. 4

Gamma-ray spectrum of an iron foil after irradiation for 4 h with protons of 12.6 MeV energy at an ion current of 5 μA. The main isotope present is 56Co

Following the foil irradiation, the pure fraction of 56Co was produced using the dissolution and separation procedure described above. The Fe ion impurity level in the 56Co radiotracer solution and 56Co-radiolabelled SiO2 NPs was determined by inductively coupled plasma mass spectrometry (ICP-MS). The results obtained are shown in Table 3. The content of Fe ions in 56Co radiotracer solution was less than 5 %, which indicates efficient 56Co/Fe radiochemical separation.
Table 3

Summary of results of Fe determination in 56Co solution and 56Co-radiolabelled SiO2 NPs by ICP-MS

  

Concentration

Sample type

Nominal (μg/L)

Determined (μg/L)

Obtained value of Fe in 56Co–SiO2-NP suspension in g/3 mL

ICP-MS calibration

Blank

0.0

STD 1

5.0

4.1

STD 2

50.0

48.7

STD3

500.0

500.1

Fe-impurity determination

56Co solution (1)/56Co–SiO2 20 nm

300/250

13.7/6.7

0.00080

56Co solution (2)/56Co–SiO2 40 nm

300/250

13.5/5.4

0.00065

56Co solution (3)/56Co–SiO2 100 nm

300/250

14.1/8.3

0.00099

The ICP-MS was calibrated with blank and three concentrations of the Fe standard solutions STD 1, STD 2, and STD 3. Typically, the nominal calibration concentrations were 5, 50 and 500 μg/L, respectively. The analysis was done in 0.5 % HNO3. In order to correct for matrix effects due to the different sources of samples and calibration standards 1,000 ppb of Sc was added as internal standard to all samples, blank specimens and calibration standards. The determination of Fe impurities was based on the assumption that all 30 mg Fe would still be in the Co-solution after eluation. Diluting the ICP-MS samples to (nominally) 300 ppb under this assumption shows that only about 14 ppb can be detected, which means that more than 95 % of the Fe is retained by the separation procedure. Applying the same procedure after NP synthesis to the NP suspensions diluted (to nominally) 250 ppb reveals impurity levels around 5–8 ppb. Calculating back to the initial specimen volume of 3 mL for 56Co–SiO2-NPs gives the ‘obtained value’ expressed in g/3 mL

DLS results of the synthesis of both non-active and 56Co-radiolabelled silica NPs are shown in Table 4. Tests were performed in MilliQ-water at the final concentration of 100 μM. The stability of the particles with respect to agglomeration was investigated by measuring the particle diameter over time in MilliQ-water and PBS solution at 24 h after synthesis and at 144 h after synthesis. Results for the SiO2 and 56Co–SiO2 NPs are presented in Table 4.
Table 4

DLS analysis of SiO2 and 56Co–SiO2 NPs in MilliQ-water and PBS solution (24 h, 144 h)

Samples

Polydispersity index ± SD

Average particle diameter (nm) ± SD

DLS analysis in MilliQ-water

 20 nm SiO2 24 h

0.029 ± 0.007

19.4 ± 0.2

 40 nm SiO2 24 h

0.034 ± 0.005

41.0 ± 0.1

 100 nm SiO2 24 h

0.057 ± 0.002

100.9 ± 0.9

 20 nm SiO2 144 h

0.030 ± 0.001

20.2 ± 0.3

 40 nm SiO2 144 h

0.037 ± 0.001

42.1 ± 0.9

 100 nm SiO2 144 h

0.079 ± 0.017

101.4 ± 0.8

 20 nm 56Co–SiO2 24 h

0.128 ± 0.036

20.6 ± 0.4

 40 nm 56Co–SiO2 24 h

0.116 ± 0.027

42.6 ± 0.8

 100 nm 56Co–SiO2 24 h

0.138 ± 0.016

103.3 ± 0.9

 20 nm 56Co–SiO2 144 h

0.165 ± 0.023

21.2 ± 0.9

 40 nm 56Co–SiO2 144 h

0.124 ± 0.020

43.5 ± 0.7

 100 nm 56Co–SiO2 144 h

0.155 ± 0.029

102.9 ± 0.7

DLS analysis in PBS solution

 20 nm SiO2 24 h

0.042 ± 0.008

20.2 ± 0.1

 40 nm SiO2 24 h

0.054 ± 0.011

40.8 ± 0.1

 100 nm SiO2 24 h

0.071 ± 0.003

99.1 ± 0.5

 20 nm SiO2 144 h

0.040 ± 0.002

19.8 ± 0.3

 40 nm SiO2 144 h

0.054 ± 0.011

41.5 ± 0.7

 100 nm SiO2 144 h

0.061 ± 0.013

100.9 ± 0.4

 20 nm 56Co–SiO2 24 h

0.146 ± 0.032

21.7 ± 0.4

 40 nm 56Co–SiO2 24 h

0.112 ± 0.016

43.4 ± 1.0

 100 nm 56Co–SiO2 24 h

0.159 ± 0.020

101.8 ± 0.9

 20 nm 56Co–SiO2 144 h

0.177 ± 0.037

22.0 ± 0.8

 40 nm 56Co–SiO2 144 h

0.142 ± 0.021

42.8 ± 0.7

 100 nm 56Co–SiO2 144 h

0.168 ± 0.028

103.5 ± 0.5

The results shown in Table 4 demonstrate that it has been possible to radiosynthesise 56Co–SiO2 NPs with a similar size as that achieved with the conventional non - active synthesis method for SiO2 NPs, under the same synthesis conditions. Most likely the minor presence of Co isotopes and remaining Fe impurities within the silica NPs contributed to a slight increase of the average nanoparticle size in comparison to those obtained under the equivalent experimental parameters without 56Co. By carefully controlling the synthesis conditions and the TEOS and ammonia concentrations, it was possible to synthesise 56Co-labelled SiO2 NPs from a size of 20 nm up to 100 nm in reproducible way.

DLS generally will indicate a significant level of polydispersity, even when the sample consists of a very monodisperse analyte. In general, samples with PdI < ~0.2 may be assumed to be quite monodisperse. The performed DLS measurements of the silica suspensions in MilliQ-water resulted in PdI numbers in agreement with good monodispersity. Both NP types (non-active and radioactive) are stable over time in aqueous suspension. As presented in Table 5, for both SiO2 and 56Co–SiO2 NPs samples, the analysed Zeta potential values confirm the stability of silica particles in aqueous medium and indicate no significant modification in the surface charge of the NPs caused by the radiolabelling process.
Table 5

Zeta potential values of SiO2 and 56Co–SiO2 NPs in MilliQ-water, pH = 7

NPs sample

Zeta potential (mV)

20 nm SiO2

−43.6 ± 1.3

40 nm SiO2

−44.2 ± 1.3

100 nm SiO2

−44.9 ± 1.2

20 nm 56Co–SiO2

−42.5 ± 1.0

40 nm 56Co–SiO2

−43.2 ± 0.8

100 nm 56Co–SiO2

−43.7 ± 0.9

Stability studies of the 56Co radiotracer within SiO2 NPs were performed in MilliQ-water and PBS solution (for 14 days at 25 °C) by the ultracentrifugation process to separate particles from the liquid medium and measure the activity remaining in the liquid. Leaching results in both media have shown negligible release of 56Co (Co2+) from the NPs. For aqueous and PBS supernatant, <0.5 % of 56Co activity was found by gamma spectrometry analysis for each size of NPs. The simple centrifugation experiments performed here confirmed that practically all the 56Co was bound into the nanoparticles during synthesis, and the radiolabels were stably incorporated into the SiO2 NPs.

The determination of the structure of the non-active and the radiolabelled 56Co–SiO2 NPs was performed by X-ray diffraction patterns. The results are presented in Fig. 5. The patterns indicate that the SiO2 synthesised with and without added 56Co had basically the same amorphous structure, which matched well typical XRD patterns of amorphous SiO2 (Zhang and Fan 2012).
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-012-1185-x/MediaObjects/11051_2012_1185_Fig5_HTML.gif
Fig. 5

XRD patterns of SiO2 and radiolabelled 56Co–SiO2 NPs

Conclusions

A simple method for radiochemical synthesis of 56Co-radiolabelled SiO2 NPs using cyclotron-generated radioactive precursor material, and a well-known SiO2 NP synthesis procedure, has been demonstrated. SiO2 cannot be effectively radiolabelled by either neutron or light-ion activation, and there are no suitable Si or O radioactive precursors available. The method opens up new possibilities for tracing studies related to nanoparticle safety. Silica NPs were chosen for this study due to their industrial and scientific interest, but the general method of introducing minute amounts of extrinsic radiotracers into a synthesis process should also be applicable to other NP types, particularly if the radiotracers will be easily chemically incorporated into the NP matrix. Special attention has been dedicated to the physico-chemical characterisation of synthesised non-active and radioactive 56Co–SiO2 NPs to ensure that the introduction of the radioisotopes, separated from cyclotron target material, did not significantly modify the synthesis process.

Acknowledgments

The author would like to express their sincere gratitude to D. Gilliland, JRC European Commission, IHCP (NBS Unit) Ispra for helpful discussions during the conduct of the experimental research and S. Fortaner, JRC European Commission, IHCP (ECVAM Unit) Ispra, Italy for his help and support in ICP-MS measurements. Part of the study has been supported by the European Commission’s 7th Framework Programme projects ‘NeuroNano’ under contract NMP4-SL-2008-214547, and QNANO under contract SP4-CAPACITIES-2010-262163.

Copyright information

© Springer Science+Business Media B.V. 2012