Covalent Functionalization for Multi-walled Carbon Nanotube (f-MWCNT)-Folic Acid Bound Bioconjugate

MATERIALS AND METHODS

Chemicals: The hydroxyamines and diamines were of analytical grade obtained from Adrich Chemie, Fluka AG, South Africa. Other solvents and reagents obtained from Adrich Chemie, Fluka AG, South Africa include: D,L aspartic acid, phosphoric acid, 3-Dimethylamino-1-Propylamide (DMP), Diethylenetriamine (DET), 2-2- (Ethylenedioxy)-Diethylamine (EDDA), 1,3 Diaminopropane (PDA), Dicyclohexylcarbodiimide (DCC), 2-(1H-Benzotriazol-1-yl)-1,1,3,3-Tetramethyluronium Fluorophosphates (HBTU), Triethylamine (TEA), ferrocene, folic acid, methotrexate, sodium hydroxide, Calcium chloride, chloridric acid, folic acid glacial acetic acid, ammonium hydroxide, sulphuric acid and nitric acid. Distilled water was used for all preparative work. The reaction solvent, N,N-Dimethylformamide (DMF), was distilled under reduced pressure with a fore-runs of around 10% being discarded and was dried over molecular sieves 4 D. All other solvents, Diethyl ether (Et₂O), hexane, acetone and toluene were of laboratory grade, received from Adrich Chemie, Fluka AG, South Africa. The acetylene, argon and nitrogen gases were also of analytical grade obtained from Afrox South Africa for the CNT production.

The production multi-walled carbon nanotube (MWCNTs): The CNTs were synthesized by the Chemical Catalytic Vapour Deposition (CCVD) which is made up of a vertical silica plug flow reactor inside a furnace. It is a modification of the chemical vapour deposition (Anreddy et al., 2010). The furnace is connected to a swirled coiled mixer with a supply of gas system of valves and rotameters. The upper part of the reactor is connected to a condenser which leads to two delivery cyclones where the CNTs are collected. The system is also connected to a temperature regulator and a pressure controller. The valves control the flow of gases into the reactor. The CNTs were produced according to Yah et al. (2011) and viewed using the JEOL JEM-100S Transmission Electron Microscopy (TEM).

Functionalization of CNTs with sulphuric acid and nitric acid: The pristine MWCNTs were treated with hydrochloric acid, to remove impurities. One gram of MWCNTs was placed in a 500 mL round-bottom flask and 200 mL of HCL (30%) was added. The mixture was stirred using magnetic stirrer for 2 h, then diluted in water, filtered washed with deionised water and then dried overnight in a vacuum at 40°C to remove the iron impurities.

To functionalize the purify CNTs, 0.1 g of the purified MWCNTs were dispersed in 200 mL of acid (mixture of sulphuric acid 95% and nitric acid 95% of ratio 3:1) in a 500 mL round bottom flask equipped with a condenser. The dispersion was kept differently for 4 h at 100°C, 24 h at 50°C and 96 h at Room Temperature (RT). The resulting dispersion was diluted in water to neutral pH, filtered and the sample was dried in a vacuum at 40°C overnight. The modified CNTs (termed f-MWCNTS) were quantitatively analyzed by titration to determine the carboxylic acid (–COOH) concentrations on the surface of treated CNTs. The f-MWCNTs were added into a 25 mL 0.04 M NaOH solution and stirred for 48 h to allow the solid CNT material to equilibrate with the NaOH solution. The mixture was then titrated with a 0.04 M HCl solution to determine the excess NaOH in the solution and the concentration of the carboxylates on CNTs.

Functionalization of CNTs with Aspartic acid: Approximately 0.15 g of MWCNTs were dispersed in a mixture of 3 g of Aspartic acid and 15 mL of N, N-Dimethyl Formamide (DMF) in a round-bottom flask of 250 mL equipped with a condenser and the reaction was controlled to 220°C for 6 h. After filtration, the solid was washed with water to neutral pH. The yield obtained was 183 mg.

Using titration formula:

(1)

where, Y = mmol of COOH on the surface of CNTs, X = The volume of HCl in mL.

The 1H NMR (29) analyses with D₂O +NaOH showed CH₂-COOH pick at chemical shift (3-2.3) expected. The TEM analysis were done according to the microscopic image obtained in Fig. 4 with nanotubes structure at d₀ = 60 nm.

Functionalized carbon nanotubes (f-CNTs)-folic acid conjugates
A: f-CNTs (H₂SO₄+HNO₃ at 100°C) bound folic acid: Approximately 100 mg of f-CNTs (prepared at 100°C) representing 0.38 mmol COOH was dispersed in 5 mL of DMF and 11.6 mg (0.114 mmol) of 3-Dimethylamino-1-Propylamine (DMP) representing 30% of molecular rate was dissolved in 4 mL of DMF. After nitrogen flashing, the dispersion and the solution were mixed together in a round bottom flask, stirred at 50°C. While solution was still stirring, 75 mg (0.32 mmol) of 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium fluorophosphates (HBTU was added and the pH was controlled to 7 using the triethylamine (TEA) dropwise. After 24 h of reaction, a solution of 141 mg (0.32 mmol) 20% excess of folic acid in 10 mL of DMF was prepared for 1 h and was added in the main solution. Stirring was maintained for 6 h while the pH was kept to 7 with TEA. The solid was collected by filtration and washed several times with Tetramethylurea (TMU) to remove the excess FA and acetone. The solid was dried and kept in the desiccators.

B: f-CNTs (Aspartic acid) bound folic acid (f-CNTs-DMP (40)-FA (60)): Also 100 mg of f-CNTs representing 40 mg (0.88 mmol) of CNT-COOH was dispersed in 5 mL of DMF and 36 mg (0.352 mmol) of 3-dimethylamino-1-propylamine (DMP) (40%) was dissolved in 4 mL of DMF. After nitrogen flashing, the dispersion and the solution were mixed together in a round-bottom flask, stirred at 50°C. During stirring, 206 mg (0.88 mmol) of 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium fluorophosphates (HBTU) predissolved in 3 mL DMF was added dropwise and the pH was neutralised to 7 using the Triethylamine (TEA). After 2 h of reaction, a solution of 303 mg (0.7 mmol) of FA in 15 mL of DMF was prepared for 1 h and was added into the main solution. Stirring continued for 6 h while the pH was kept at 7 with TEA. The solid was washed several time with Tetramethylurea (TMU) and filtered to remove the excess FA and acetone. The solid was dried and kept in the desiccators.

The ¹H NMR spectra and the Solid-state Infrared (IR) analysis: The ¹H NMR spectra were obtained at 300 and 400 MHz in DO₂ solution. The chemical shifts, δ in ppm were referenced against sodium 3-(trimethylsily)-2, 2, 3, 3-d₄ propionate. The pH values of the sample in DO₂ solution were adjusted to 10-11 by adding sodium hydroxide in order to eliminate potential protonation effect. The Solid-state Infrared (IR) spectra were recorded on KBr pellets over the region of 4000-6000 cm^-1. The attention was turned only on the significant bands. Samples for analysis were dried using the Abderhalden apparatus and calcium chloride was done on a HITACHI 2000 spectrometer, at a scan speed of 400 nm min^-1.

RESULTS AND DISCUSSION

Synthesis of carbon nanotubes: The analysis of the Chemical Vapor Deposition (CVD) product by TEM showed MWCNTs with outer diameter (d_o) of 20 nm and an inner diameter (d_i) of 10 nm as shown in Fig. 1. The MWCNTs were similar to those earlier produced and reported by Iyuke et al. (2009) and Mohamed and Kou, 2011 where the optimal temperature for production was found at 900⁰C with similar diameter. The results were dissimilar from those produces by Mamba et al. (2010) where they found MWCNTs as straight bundles of uniform lengths, each bundles consisting of numerous closely packed MWCNTs ropes with no indication of surface modifications.

The CNTs found could have a variety of applications because of their novel structure, especially the small size and remarkable mechanical, thermal and electrical properties (Zambri et al., 2011). Small-diameter MWCNTs have ability to display enhanced reactivity relative to larger-diameter nanotubes, due to increased curvature strain. However, because of their novel structure of elemental carbon, CNTs are insoluble in all organic solvents (Kostarelos, 2003).

Covalent functionalization of carbon nanotubes
Functionalization of CNTs with concentrated nitric acid and sulfuric acid: (a) The MWCNTs were functionalized with H₂SO₄/HNO₃ in mol ratio of 3:1 at 100, 50°C and Room Temperature, respectively (RT). The TEM analysis of the f-MWCNTs showed variation at 100°C with outer diameter (d₀) = 20 nm and the inner diameter (d_in) = 12 nm (Fig. 2) while at 50°C d_o= 30 nm (Fig. 3) and at Room Temperature (RT) d₀ = 80 nm and d_in= 40 (Fig. 4). In order to evaluate the variation of the size of f-CNTs in terms of temperature, a graph of diameter versus temperature was plotted as shown in Fig. 5. The water solubility of f-CNTs was decreasing from 100°C to RT especially from pH = 9 and above.

The analysis of ¹H NMR Spectra 1, 2, 3 done for f-CNTs obtained at 100°C, 50°C, RT respectively showed peaks A and B in chemical shift δ/ppm (1.5-2.5) for –CH₂COOH and (2.7-3.7) for –CH₂OH. In order to calculate the quantity of COOH and OH, the integration of peak A (–CH₂COOH) corresponding to 1 mol and was used as reference to evaluate the peak B (–CH₂OH). After calculation, the following results were obtained:

1H NMR Spectra 3:	f-CNTs: MWCNTs in H2SO4/HNO3 at RT

Fig. 1:	MWCNTS microscopic image (TEM)

Fig. 2:	f-CNTs (H2SO4/HNO3 at 100°C) microscopic image

Fig. 3:	f-CNTs (H2SO4/HNO3 at 50°C) microscopic image

Fig. 4:	f-CNTs (H2SO4/HNO3 at RT) microscopic image

Fig. 5:	Size of CNTs versus temperature

H NMR Spectra 1:	f-CNTs: MWCNTs in H2SO4 / HNO3 at 100°C

1H NMR Spectra 2:	f-CNTs: MWCNTs in H2SO4/HNO3 at 50°C

A sample of 200mg of f-CNTs was used for titration to evaluate the quantity of COOH in mmol incorporated on the surface of f-CNTs which through the mol ratio COOH/OH was calculated from different ¹H NMR spectra. The integration was used to calculate the quantity of OH in mmol incorporated on the surface of f-CNTs. The experimental data from titration was used through the formula Y = 1-0.04 X to complete Table 1. The results in Table 1 were used to plot fraction of incorporation of COOH and OH versus the temperature as shown in Fig. 7 and the mol ratio COOH/OH versus the temperature as shown in the Fig. 8.

Different peaks including A, B, C, D and E were shown in the infrared (IR) investigation as presented in Fig. 6.

Fig. 6:	Infrared Peaks of f-CNTs (H2SO4/HNO3) at 100°C

Fig. 7:	Fraction of COOH and OH group incorporation versus temperature

The results in Fig. 5 are elaborated from the TEM analyses in Fig. 2-4 which showed increase in size of nanotubes which were relative to decrease in temperature. This might have been due to self-diffusion of molecules inside of CNTs thereby allowing the CNTs to form a well-ordered nanoporous membrane (Casavant et al., 2003; Hinds et al., 2004) which can be incorporated in a macroscopic structure (Srivastava et al., 2004) for separation devices. The study of self-diffusivity can explain the non parabolic influence of nanotube sizes and temperature. Jobic and co-workers had examined this effect in more detail for diffusion (Jobic et al., 2009) using molecular dynamics simulations.

Fig. 8:	Mol rate COOH/OH versus temperature

Table 1:	Fraction incorporation of COOH and OH on the surface of CNTs with variable temperature
2nH: Number of Hydrogen atom, 3X: The volume of HCl in mL, 4Y: The quantity of COOH in mmol incorporated on the surface of f-CNTs

They termed this behaviour the “Levitation effect”. If the guest molecule fits perfectly in the window of the zeolite, the molecule appears to be floating; increasing or decreasing the diameter of the guest molecules as well as reducing the diffusion coefficient. This effect is temperature dependent (Derouane et al., 1988). A similar effect has been observed by Cannon and Hess (2010) for carbon nanotubes.

With regard to Fig. 5, the self-diffusivity was subsequently increasing with decreasing of temperature. Since the self-diffusion explains the incorporation of guest molecule in CNTs through nanoporous membrane, the diameter of CNTs has to increase in the tendency where the self-diffusion will increase. According to the objective of the study, attention was given to the trend where the surface adsorption (surface phenomenon) was possible, rather than where absorption was evident. The increase temperature which led to decrease in diameter, resulted in higher collision frequencies and faster motion. This enhances the particle probability to bond folic acid. The f-MWCNTs at 100°C were selected, due to their small size and the adsorption capabilities.

The ¹H NMR Spectra 1, 2 and 3 were obtained with D₂O as solvent. Therefore COOH and OH groups directly attached on the surface of CNTs were shifted in δ/ppm (4.5-5.5) for D₂O peaks.

Scheme 1:	f-CNTs (H2SO4/HNO3) structure

Scheme 2:	f-CNTs (aspartic acid) structure

The ¹H NMR analysis for all f-CNTs made at different temperatures (100, 50°C and RT) shown the presence of CH₂COOH and CH₂OH due to the fact that amorphous carbon was removed purification by HCL (Monthioux et al., 2010).

In addition, the infrared analysis from Fig. 6 and the ¹H NMR Spectra 1, 2 and 3 have confirmed that f-CNTs were produced and the entire MWNTs was covered with CH₂OH, CH₂COOH, OH and COOH as shown in Scheme 1.

In comparison with the result from Fig. 7 and 8, it was found that both fractions of OH and COOH incorporation in f-CNTs were increasing with increasing temperature. At room temperature, the fraction of OH group incorporation was much neglected and the mol ratio COOH/OH was 32.4, these were rapidly reduced to 1.1 at 50°C and sharply down to 0.35 at 100°C. This shows that when working at room temperature or below there is possibility of getting f-CNTs covered only with COOH. It was found that f-CNTs at 100°C were more soluble than at RT and 50°C due to high presence of OH groups on the CNTs surfaces. Thus, the abundance of OH and COOH groups on the surface will facilitate more covalent attachment to polymers or drugs according to their different electronegativities. This granted the f-CNTs made at 100°C to be more useful than those made at low temperatures. The scope of this reaction was very broad and produces f-CNTs that possess high solubility in a wide range of different solvents. This was preferentially selected for the next phase of the drug binding to carbon nanotubes.

Where Series 1: Fraction incorporation of carboxyl group (COOH) on the surface of CNTs and Series 2: Fraction incorporation of phenol group (OH) on the surface of CNTs.

Functionalization of CNTs with Aspartic acid: The functionalized carbon nanotubes (f-CNTs) were also obtained by heating MWNTs with aspartic acid an amino acid in DMF at 220°C. This resulted to the structure as shown in Scheme 3. The results of f-CNTs by aspartic are shown in Fig. 9 where 183 mg of f-CNTs by aspartic acid were soluble in water at pH of 10.

The ¹H NMR Spectra 4 showed peaks A in chemical shift (δ/ppm) 3.050-2.200 for –CH₂COOH and B in chemical shift (δ/ppm) 4.40-4.70 for = N-CH CH₂ (COOH)₂ as expected. The different peaks was obtained from infrared (IR) investigation (A, B, C, D, E, F and G) are shown in Fig. 10.

Using the titration formula Y = 1-0.04 X (Y: quantity of COOH in mmol incorporated on the surface of f-CNTs and X: volume of HCl in mL) when working with 200 mg of f-CNTs 0.88 mmol of COOH corresponding to 20% of COOH was incorporated on the surface of CNTs after neutralization with excess NaOH and 3 mL of HCl. The carboxyl group incorporated in the f-CNTs was intended to be bound with the solubilizing group and the drug. The quantity of the solubilizing group and drug were evaluated with 20% of fraction of COOH-f-CNTs at pH of 10. The TEM analysis showed f-CNTs-aspartic acid diameter at 60 nm which was larger than the diameter obtained for MWNTs in sulfuric and nitric acids at 100 and 50°C and to a lesser extend at RT. According to the findings from Fig. 5, the diameter was decreasing with increase in temperature which was similar to those earlier reported by Hunger et al. (2002), Jakobtorweihen et al. (2006). The investigation of peaks in ¹H NMR Spectra 4 and infrared (IR) as shown in Fig. 10 which confirm the structure elaborated in Scheme 2.

Fig. 9:	f-CNTs (MWNTs +Aspartic acid in DMF at 220°C)

Fig. 10:	IR f-CNTs (aspartic acid)

Overall, the use of strong acid and amino acid to functionalize the carbon nanotubes generates carboxylic groups (Scheme 1, 2) which increase their dispersibility in aqueous solutions (Tasis et al., 2006; Kim et al., 2007). Solubility under physiological conditions is a key prerequisite to make CNT biocompatible which can be linked to a wide variety of active molecules.

Functionalized carbon nanotubes (f-CNTs) bound Folic acid conjugates: The folic acid was functionalized with the MWCNTs by the aid of 3-dimethylamino-1-propylamine (DMP) as a strong base. This gave the amino group a positive charge enabling the FA to bind. This was similar to those reported by Zeng et al. (2006) were they prepared functionalized CNTs with biodegradable poly (e-caprolactone) (PCL). They found that the presence of CNTs did not affect the biodegradability of PCL. From HNMR results it was found that FA was attached to surface of the MWCNTs. Furthermore functionalized CNT-PCL polymer matrix could display enormous potential as a structural support material in tissue engineering.

f-CNTs (H₂SO₄+HNO₃ at 100°C) bound folic acid conjugates: The estimation of the percentage of -COOH in f-CNTs was proved by the titration. The f-CNTs was achieved by the reaction between –COOH and NH₂ via the coupling agent 0-benzotiazol-1-yl-tetramethylurorium hexafluorophosphate (HBTU). Thus, the coupling led to the yield of 163 mg f-CNT-FA similar to the chemical structure presented in Scheme 3. The TEM analysis of f-CNTs (H₂SO₄+HNO₃ at 100°C) bound folic acid conjugates are shown in Fig. 11 with d₀ = 50 nm and d_in = 20 nm and soluble at pH 6-7. The ¹H NMR Spectra 5 were similar to peaks shift shown in Scheme 3 as illustrated in Table 2.

With the size of 50 nm as outer diameter and 20 nm as inner diameter the conjugate can be easily tracked into the cytoplasm and the nucleus. With regard to the ¹H NMR Spectra 5, the fraction of incorporation of folic acid was calculated by using the integration of A that is 1 and the sum of integration of aromatic group E, F and G which is 1.64. With 30% of DMP and 70% of folic acid, the mol ratio was 3/7 and the number of H expected was 20 H for A and 35H for the sum of E, F and G as it is shown in the structure (Scheme 3). Referring to A as 20H calculated with integration 1, the sum of E, F and G with integration of 1.64 gave 32.8H calculated which corresponded to 94% incorporation.

f-CNTs (Aspartic acid) bound folic acid conjugates: The estimation of the percentage of –COOH in f-CNTs was also evaluated by the titration which was achieved by the reaction between –COOH and NH₂ via the coupling agent 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium fluorophosphates (HBTU).

Scheme 3:	f-CNTs (H2SO4/HNO3)-DMP (30)-FA (70) Structure

Scheme 4:	f-CNTs (Aspartic acid)-DMP (40)-FA (60) Structure

Fig. 11:	f-CNTs (H2SO4/HNO3)-FA

Table 2:	HNMR spectra of f-CNTs (H2SO4/HNO3)-FA
1Y is the quantity of COOH in mmol incorporated on the surface of f-CNTs and X is the volume of HCl in mL

Table 3:	HNMR data f-CNTs (aspartic acid)-DMP (40)-FA (60)
1Y is the quantity of COOH in mmol incorporated on the surface of f-CNTs and X is the volume of HCl in mL

With the objective to raise the water solubility of f-CNTs, 40% of –COOH was evaluated using attached DMP as the solubilizing group. The folic acid was used in excess with the objective to complete all the 60% of -COOH that did not react with DMP. The result f-CNTs (Aspartic acid) bound folic acid conjugate yielded 231 mg and the chemical structure is presented in Scheme 4. The TEM analysis is shown in Fig. 12 having diameter of 170 nm. The product was soluble at pH of 6-7.

The ¹H NMR Spectra 6 showed all expected chemical bands as illustrated in Table 3. This therefore, confirms the reversible attachment between f-CNTs with DMP and folic acid via amid bond formation. To calculate the fraction incorporation of folic acid, the integration of peak A evaluated to 1 was chosen as reference and corresponded to 4 H as expected and the integration of aromatic peaks (E, F, G) evaluated to 3.8 was calculated 15.2 H while the expected was 15 H as shown in Scheme 4 and then the fraction incorporation of folic acid was 101.3%. The error was 1.3%.

The f-CNTs-drug conjugate from aspartic acid (Fig. 12) showed the dark spot spread along the nanotubes while Fig. 11 related to the f-CNTs of conjugate made from H₂SO₄/HNO₃ at 100°C also dark spot at the end cap.

Fig. 12:	f-CNTs (Aspartic acid)-DMP (40)-FA (60) microscopic image

1H NMR Spectra 4:	f-CNTs: MWCNTs+Aspartic acid in DMF at 230°C.

1H NMR Spectra 5:	f-CNTs (MWCNTs+H2 SO4+HNO3)-DMP (30)-FA (70) 94% FA incorporation

The incorporation of COOH (17.1%) in f-CNTs with H₂SO₄/HNO₃ at 100°C (Table 1) was lesser than the one (20%) in f-CNTs with aspartic acid.

1H NMR Spectra 6:	f-CNTs (MWCNTs+Aspartic Acid)-DMP(40)-FA(60) Incorporation FA 101,3%

CONCLUSION

The carbon nanotubes, due to its small size were synthesised as drug carrier target. The covalent functionalization was developed on the surface of MWCNT. The carboxyl groups reacted with HBTU as coupling agent DMP amines in one part and folic acid in other. The formation of CONH as bioreversible bond between the f-CNTs and folic acid was determined by ¹H NMR and IR. The soluble f-MWCNTs TEM structures were conserved within the diameter range of 50 to 170 nm. The phenol (OH) and carboxyl (COOH) groups were attached on the surface of CNT making the rate COOH/OH to decrease with increasing temperature. The f-CNT covered surface with OH and COOH was more water soluble than the f-CNT covered COOH.

ACKNOWLEDGMENTS

The authors acknowledge the financial support from the National Research Foundation (NRF) under the NRF Focus Area: NRF Nanotechnology flagship programme, DST/NRF Centre of Excellence. The student bursaries provided by the Wits University are much appreciated.