Metallized Carbon Nanotubes for Radiation Therapy Enhancement

Background: Carbon nanotubes have been the focus of many in-vitro and in-vivo studies where they serve as cargo delivery vehicles. In this preliminary work, multi-walled carbon nanotube were impregnated with copper, to study the effect of metallized carbon nanotubes on enhancing radiation dose in external beam radiotherapy. Methods: Using a 6x6x6 cm 3 cube, lled with water or with water plus partially lled or unlled carbon nanotubes, the dose enhancement effects of megavoltage photon beams were determined. Ion chamber readings were collected at 1 cm below the water cube. The beams were delivered from a TrueBeam Linac, at a 100 cm source to surface distance and 20x20cm 2 eld size. In addition, lm measurements were taken just below the water cube (i.e. at 0 cm distance) and compared to the ion chamber dose enhancement at 1 cm below, for 6 MV energy beam. Results: For all studied photon energies, the ion chamber readings showed a dose increase when copper lled carbon nanotubes were added to the water medium. With copper lled CNTs concentration in water of 5 x 10 -2 mg/ml, the measured dose enhancement ratios (DER) were 1.025, 1.016, 1.017, 1.011 and 1.013for 2.5MV, 6MV, 6MV FFF, 10MV and 10MV FFF energies, respectively. Furthermore, when copper lled MWCNTs are added to water, the lm measurements (at 0 cm distance from the water cube) have also shown a dose enhancement ratio of 1.022 in comparison to 1.012 for ion chamber (1 cm away from the water cube). Conclusion: The results show an increase of the radiation dose enhancement, with a DER up to 1.025 at 1 cm below the studied medium, when metallized carbon nanotubes are added to water versus when the beam is delivered to just water. This increase in DER decreases with the increase of x-ray beam energy. It also decreases with increasing the distance of the point of measurement from the medium where the metalized carbon nanotubes were added.

Results: For all studied photon energies, the ion chamber readings showed a dose increase when copper lled carbon nanotubes were added to the water medium. With copper lled CNTs concentration in water of 5 x 10 -2 mg/ml, the measured dose enhancement ratios (DER) were 1.025, 1.016, 1.017, 1.011 and 1.013for 2.5MV, 6MV, 6MV FFF, 10MV and 10MV FFF energies, respectively. Furthermore, when copper lled MWCNTs are added to water, the lm measurements (at 0 cm distance from the water cube) have also shown a dose enhancement ratio of 1.022 in comparison to 1.012 for ion chamber (1 cm away from the water cube).
Conclusion: The results show an increase of the radiation dose enhancement, with a DER up to 1.025 at 1 cm below the studied medium, when metallized carbon nanotubes are added to water versus when the beam is delivered to just water. This increase in DER decreases with the increase of x-ray beam energy. It also decreases with increasing the distance of the point of measurement from the medium where the metalized carbon nanotubes were added.

Background
Several studies have focused on the use of metallic nanoparticles for enhancing the radiation dose where the target lesions are located, while sparing the surrounding healthy tissue 1,2,3,4 . Gold is the most-studied nanoparticle for dose enhancement, while very little has been published on the use of metallized carbon nanotubes as radiation dose enhancers in external beam radiotherapy. Carbon nanotubes (CNTs) are a carbon allotrope with unique physical and chemical properties. Due to their needle-like shape, they have superior cell permeability compared to other types of nanoparticles and a large aspect ratio, allowing a large surface area for functionalization and modi cation 5,6,7 . They can also be shortened and functionalized to decrease toxicity and enable chemical speci city towards targeted sites. In addition, both the hollow interior and the outside of the CNTs can be utilized to carry treatment and targeting agents simultaneously. Because of these considerations, carbon nanotubes have been the focus of many in-vitro and in-vivo studies where they serve as cargo delivery vehicles. In this preliminary work, the radiation enhancer of choice is copper, encapsulated in the multi-walled carbon nanotube (MWCNTs).
This choice was made due to the cost effectiveness of copper salt such as Copper (II) Nitrate Trihydrate compared to gold analogs. Copper nanoparticles, however, have shown liver toxicity in rats due to oxidation 9 . Encapsulating the copper within carbon nanotubes walls can preserve it from oxidation 10 . In addition to the effect of metalized carbon nanotubes, the effect of pure MWCNTs was also studied Methods MWCNTs with > 98% purity were obtained from Sigma-Aldrich (Sigma-Aldrich, Saint Louis, MO). The MWCNTs possessed outer diameters of 6-13 nm and lengths of 2-20 µm.
The MWCNTs were opened and lled with copper in a two-step process 8 . 500 mg of the nanotubes were rst re uxed in nitric acid ( Fig. 1-a). When re uxing, the MWCNTs and the acid were combined in a ask on a hot plate with a spinning magnetic stirring rod. As the acid boils, the vapor rises and condenses in the re uxing tube, preventing it from boiling away. At the same time, the MWCNTs are being stirred in the hot acid. The heat, along with the mechanical force, breaks the carbon-carbon double bonds at the ends of the tube and simultaneously functionalizes the walls with an OH group ( Fig. 1 To examine the outcome of the partial lling process, scanning electron microscope (SEM) and transmission electron microscope (TEM) images were acquired along with energy dispersive x-ray spectroscopy (EDS) and x-ray photoelectron spectroscopy (XPS) data. SEM gives morphological information of the samples, EDS and XPS give chemical information and elemental composition, TEM veri es that the copper nanoparticles are located within the nanotubes and not attached to the outside walls.
For the dose enhancement study, we have added 10 mg of Cu-MWCNTs to 216 g of water, with < 1 mg of copper in the water cube. We studied the effect of adding un lled and Cu-lled MWCNTs to an aqueous medium on the measured radiation dose. A Varian TrueBeam Linac (Varian, Palo Alto, CA), was the source of the megavoltage x-ray radiation. A 6 × 6 × 6 cm 3 cube from the Max-HD phantom (IMT, Troy, NY) was lled with water and placed at 100 cm from the x-ray target. Another cube with the same dimensions made of solid water material (a plastic-like material with electron density close to water's electron density) was placed below the water-lled cube, such that the beam passes through the water cube rst, with radiation eld size of 20 × 20 cm 2 at 100 cm source to surface distance. This setup was chosen for its simplicity, while the eld size was to assure consistency with future measurements that may require a large eld size. The lower cube has several inserts allowing for micro ion chamber measurements.
An A16 ion chamber, with an active volume of 0.007 cc (Standard Imaging, Middleton, WI) was placed in the hole closest to the water cube, at 1 cm below the water cube; all other holes had solid water plugs inserted in them to minimize any effect of inhomogeneity (Fig. 2). After the initial readings were acquired, metallized or as-received MWCNTs were added to the water cube with concentration of 5 × 10 − 2 mg/cc, and readings were taken using the same setup within an hour from the original readings to avoid set up discrepancies and machine output variations. After the readings were complete, veri cation of the initial readings of the water lled cube with no metallized MWCNTs suspended in the medium were done by repeating the initial readings and comparing the two readings to each other.
These measurements were repeated using a Farmer-type ion chamber (active volume of 0.6 cc), placed at 1 cm below the water cube along the central axis of the radiation beam. This was to verify the effect of the off-center measurements, conducted by the A16 ion chamber, on the measured dose enhancement.
The off-center positioning was mandated by the design of the phantom, where the closest insert to the top is shifted to the side of the phantom.
Measurements of 5 different x-ray energy spectra were conducted twice, for consistency check.
In addition, a gafchromic-type lm (Ashland Advanced Materials, NJ) was sandwiched between the water cube and the lower cube, where the ion chamber is inserted. The phantom was irradiated with and measurements were taken with lm and ion chamber, simultaneously. This was done for a 6MV energy beam, with the cube being lled with water only, water plus copper lled MWCNTs and water plus pure MWCNTs. The DERs were calculated for lm and ion chamber, i.e. at 0 cm distance from the water cube and at 1 cm beneath it. This ratio is calculated for each setup, such that the DER for lm is the ratios of lm measurements and the DER for ion chamber is the ratio of ion chamber measurement.
JMP software (SAS Institute Inc., Cary, NC) is used for statistical analysis of the acquired data and the statistical signi cance is determined. We used one-way Anova with Tukey-Cramer post-hoc testing for all measurements and student's t-test for each pair of samples at a given energy.

Results
SEM images (Fig. 3, a) can be used to con rm that carbon nanotubes are collected on the substrate, but not whether a sample contains copper. However, SEM can be used in conjunction with EDS measurements. The EDS spectra (Fig. 3, b) showed the presence of copper at bright areas of the MWCNT SEM images. XPS spectra (Fig. 4)

Discussion
The copper lling of the MWCNTs is con rmed via a combination of various modalities, where EDS and XPS data showed the existence of copper within samples, and the TEM images provided proof of the lling of the MWCNTs with copper.
Following the con rmation of the copper lling of the carbon nanotubes, the effect of these Cu-lled MWCNTs on enhancing the radiation dose in a water cube is studied, in megavolt x-ray beams where the dominant type of radiation interaction in that energy range (between 2.5 and 10 MeV) is Compton scattering, an interaction between the incident photon and a free electron 11 . This type of interaction is independent of the atomic number "Z" of the absorbing material, as opposed to the photoelectric effect, which has a mass-dependent cross-section but is a less likely interaction in the studied energy range. Figures 6 and 7 show the dose increases when the MWCNTs are lled with copper, which has a higher atomic number than carbon and water equivalent, suggests that the photoelectric interaction is responsible for this dose enhancement. Both photoelectric effect and Compton scattering would be 63.55 g/mol of Cu, which is equivalent to 55 mg of Cu in the sample. The mass of Cu in a Cu-CNT sample would not exceed 10% of the total mass.
We used TEM to determine the location of the copper with respect to the MWCNTs. The TEM images such as in Fig. 5 shows that the MWCNTs are partially lled with copper.
The dose increase for each studied beam energy is plotted in the graph below (Fig. 6).
To demonstrate the speci c effect of adding copper lled nanotubes, MWCNTs with partial copper lling were compared to plain water with respect to dose enhancement.
There is a statistically signi cant (P < 0.05, Student's t-test) difference between the water doses with and without metallized MWCNTs suspended in the medium. They are statistically signi cant to better than 95% con dence, for all ve energies.
Furthermore, the effect of distance on the measured radiation dose enhancement was studied, using an EBT2 -type radiochromic lm and an ion chamber. The results are shown in Fig. 7, such that the lm measurements are averages of 3 pixel values of exposed areas in the lm to 300MUs and that the ion chamber measurement is a reading at 300 MUs, with the lm in the way. The ion chamber reading was veri ed after removing the lm.
In addition to the effect of metalized carbon nanotubes, the effect of pure MWCNTs was also studied. Figure 7, below, shows an increase in both cases at two different distances from the water cube.
The off-axis measurement was con rmed to not impact the results with measurements taken along the central axis.
decrease with increasing photon energy, the decrease in the dose enhancement shown in Fig. 6 is consistent with both types of interactions.
Furthermore, the dose increase just below the water cube was measured by placing a radiochromic lm between the cube and the solid water slab (Figs. 7 and 8). We see a dose increase with values that are different than the ion chamber measurements. When un lled MWCNTs are added to water, there's no photoelectric effect introduced, given that the density of the MWCNTs is less than the effective density of water, therefore the main increase would be due to Compton scattering. This increase is equal at two different distances from the water cube, at 0 cm and at 1 cm away. When the pure MWCNTs are replaced with copper lled ones, for the same photon energy (6 MV in this case), we notice a large increase in the proximal measurements (the lm that is placed just under the cube) and a small increase in the distal point of measurement ( the ion chamber, 1 cm below the water cube). This suggests that at the vicinity of the interaction, photoelectrons due to the high Z of the copper contribute most to the dose increase wit this effect decreasing with distance. Further measurements will be done to support this hypothesis.
The dose increase when pure carbon nanotubes were added to water (Figs. 7 and 8) can be explained by the probability of Compton scattering. Compton scattering probability is a function of the classical radius of the electron 12 , as in the Klein-Nishina formula such that: Compton cross section per unit solid angle (dσ/dΩ) α r 2 with "r" being the electron's classical radius, which is inversely proportional to the rest mass of the electron. The effective mass of the electron in carbon nanotubes was studied by Roy and Maksym 13 in units of the electron's rest mass, a function of the radius of the MWCNT. This leads to an expected increase in the probability of Compton scattering.
Adding copper lled carbon nanotubes to water would increase the measured dose, due to higher probability of photoelectric effect from the low energy photons in the spectrum. As we increase the energy, this probability decreases and so does the probability of Compton effect, therefore the increase in the measured dose (due to dose enhancement), decreases with increasing the energy of the beam. In the studied material, pair production is negligible. Also, secondary electrons in the beam react with nanoparticles in a different way than they would in bulk material (Krasheninnikov and Nordlund 2004), owing to the large surface area of the nanotubes the increased probability of interaction.
In addition, CNTs would potentially have a different effect, compared to other nanoparticles (e.g. Au-NPs) due to their unique honeycomb like shape, allowing electrons to escape with no self absorbance 16 .
Cu-MWCNTs concentration in the water cube is 5 × 10 − 2 g/cc, this is much lower than the recommended concentration of nanoparticles to achieve dose coverage at the cellular level 14 .
However, these recommendations were given to nanoparticles of sizes ranging between 2 nm and 100 nm and carbon nanotubes have diameters in this range but lengths in µm range. The metal particles are also distributed along the axis of the nanotubes. This could alter the recommended concentrations; further studies will be conducted to nd the optimum number.

Conclusion
Carbon nanotubes were chosen as vehicles due to their superior properties, including high aspect ratio, ability to utilize the hollow inside and the outside walls simultaneously, and documented ability of protecting encapsulated materials within them from oxidation 10 , thus reducing the toxicity of the copper. Future studies are planned to support this hypothesis. Copper was chosen, as ller, for novelty given that the published data on its role as a radiation dose enhancer is very limited.
Successful lling of MWCNTs with copper has been demonstrated. The exact amount of copper within the carbon nanotubes can not be measured directly at this time but an estimate was calculated. In the future, other tools will be explored for more accurate measurement of the lling material's volume or mass. Ion chamber measurements have shown that adding Cu-lled MWCNTs to an aqueous medium increases the measured dose, relative to water readings by up to 2.5%. This is a proof of concept that metallized carbon nanotubes can be used for radiation dose enhancement in external beam therapy with the effect of the dose enhancement being most signi cant at lower beam energies in the range tested.
The ndings from this study show dose enhancement, but the underlying reasons are not well-understood yet. More investigation in the effect of the unique properties that carbon nanotubes possess is in our future work. In addition, there is limited understanding of the biological effect of carbon nanotubes and their risk-bene t factor. Many investigators are actively researching this eld.
Future directions for this work include measuring the dose enhancement for variable concentrations of Cu-MWCNTs in water and comparison between simulation values of the dose enhancement vs. measured data. In addition, it's intended to study the effect when the copper lling is replaced with a higher Z material. This study addresses only the radiation dose enhancement, which is a physical property, and does not approach the radiosensitization effect. Given the importance of biological models, in the future we will study the radiosensitization of metal-lled carbon nanotubes and it's radiobioligical and clinical effects.

Declarations
Ethics approval and consent to participate: Not applicable Consent for publication: I give "Cancer Nanotechnology" permission to publish this manuscript.
Availability of data and materials: All data is available and the material is reproducable.
Competing interests: Not applicable Funding: Not applicable