Investigation on the storage of benzotriazole corrosion inhibitor in TiO₂ nanotube

Spherical commercial ${\rm Ti}{{{\rm O}}_{{\rm 2}}}$ (${\rm Ti}{{{\rm O}}_{2{\rm ,Co}}}$, 99.4%) of industrial grade was purchased from ROHA Dyechem (Vietnam) with average diameter of 100÷150 nm. 1,2,3-benzotriazole (BTA, 99%) was purchased from Sigma-Aldrich. Sodium hydroxide, ethanol, hydrochloric acid... were of analytical grade. All chemicals were used as received without any further purification.

TNT was prepared by hydrothermal treatment of ${\rm Ti}{{{\rm O}}_{2{\rm ,Co}}}$. Mixture of ${\rm Ti}{{{\rm O}}_{2{\rm ,Co}}}$ and NaOH 10 M solution (${\rm Ti}{{{\rm O}}_{{\rm 2}}}$:NaOH molar ratio of 1:23) was stirred continuously for 1 h [14]. The mixture was autoclaved for hydrothermal treatment at $150{{\,}^{\circ }}{\rm C}$ for 24 h. The solid in the autoclave was then filtered, washed with distilled water, soaked in dilute hydrochloric acid solution for 1 h and then rinsed with distilled water until neutral pH. At the end of the process, the solid was dried at $100{{\,}^{\circ }}{\rm C}$ for 12 h and then calcined at $500{{\,}^{\circ }}{\rm C}$ for 2 h with a heating rate of $2{{\,}^{\circ }}{\rm C}\,{\rm mi}{{{\rm n}}^{-1}}$. The resulting product is symbolized as TNT.

In this study, we used the vacuum impregnation and the vacuum evaporation to introduce BTA into ${\rm Ti}{{{\rm O}}_{{\rm 2}}}$ nanotubes. The processes were carried out in the dark to avoid the BTA degradation by light.

2.3.1. Method of rotary vacuum evaporation.

2.3.2. Method of vacuum impregnation.

A schema of vacuum impregnation is presented in the figure 1. An amount of 0.35 g of TNT was introduced into the glass jar, which was then sealed and vacuumed for 6 h to evacuate the air in the TNT (vacuum step). Then a solution containing 50 ml of ethanol and 0.5 g of BTA was slowly introduced and kept for 24 h (impregnation step). Due to pressure difference between the outside and the inside of ${\rm Ti}{{{\rm O}}_{{\rm 2}}}$ tube, BTA moved into the ${\rm Ti}{{{\rm O}}_{{\rm 2}}}$ tube. The TNT loaded BTA was recovered by centrifugation and dried at $80{{\,}^{\circ }}{\rm C}$ for 24 h.

Zoom In Zoom Out Reset image size

The vacuum impregnation was performed at two different temperatures which were ambient temperature and $3{{\,}^{\circ }}{\rm C}$ (controlled by ice water bath). The difference of these two methods was demonstrated at the vacuum step. At ambient temperature vacuum, it was called normal temperature impregnation (abbreviated as norm.imp). When the glass jar was placed in an ice water bath, this process was called cooling temperature impregnation (abbreviated as cool.imp). The purpose of cooling temperature impregnation was to lower the pressure in the glass and to load more BTA into TNT. The resulting products were denoted as BTA/TNT (norm.imp) and BTA/TNT (cool.imp), respectively.

Some characterizations were combined to evaluate TNT, BTA/TNT products. The morphology was examined by TEM and SEM images. The BET specific surface was measured using nitrogen isotherm adsorption on ASAP2020 (Micromeritics). The BTA storage was evaluated by TGA-DSC thermal analysis on STA6000 (Perkin Elmer) and by FT-IR spectroscopy on Nicolet iS10 (Thermo Scientific^™). Thermal analysis was executed at a temperature range of $30-600{{\,}^{\circ }}{\rm C}$ with heating rate of $10{{\,}^{\circ }}{\rm C}\,{\rm mi}{{{\rm n}}^{-1}}$ in inert ${{{\rm N}}_{2}}$ atmosphere with flow rate of 20 ml min⁻¹. FT-IR spectra were recorded by transmission mode using KBr pellets, within the $4000-400\,{\rm c}{{{\rm m}}^{-{\rm 1}}}$ range.

SEM images in figure 2 show that spherical ${\rm Ti}{{{\rm O}}_{{\rm 2,Co}}}$ powder of about 100÷150 nm diameter was transformed to the tubular ${\rm Ti}{{{\rm O}}_{2}}$ after hydrothermal treatment. The TNT structure is quite uniform.

Zoom In Zoom Out Reset image size

The size of TNT is observed clearly on the TEM image in figure 3(a). TNT has a diameter of 10÷12 nm and an average tube length of 100 nm. These images show that the hydrothermal process has successfully transformed ${\rm Ti}{{{\rm O}}_{{\rm 2,Co}}}$ into ${\rm Ti}{{{\rm O}}_{2}}$ nanotubes. This result indicates an increase of ${\rm Ti}{{{\rm O}}_{2}}$'s specific surface.

Zoom In Zoom Out Reset image size

In comparison with the TEM image of BTA/TNT (norm.imp) obtained from vacuum impregnation at ambient temperature (figure 3(b)), most of the empty space inside the tubes is no longer clearly seen. This phenomenon proves the presence of BTA in ${\rm Ti}{{{\rm O}}_{2}}$ tubes. The difference between the vacuum pressure inside the tubes and the atmospheric pressure outside the tubes created favorable conditions for the BTA moving into the ${\rm Ti}{{{\rm O}}_{2}}$ tubes.

However, the TEM image of BTA/TNT (rota.eva) product from rotary vacuum evaporation (figure 3(c)) shows the opposite. ${\rm Ti}{{{\rm O}}_{2}}$ tubes are mostly transparent, but there are many particles of larger diameter accumulated at TNT orifice. It can be seen that, in this case, BTA does not enter the tube but lies outside the tube. This phenomenon is due to the pressure difference between the outside and the inside of the tube not being great enough for BTA moving into the tube.

Results of the BET surface measurements by the isotherm adsorption on ASAP2000 are presented in table 1. The specific surface area of TNT increases significantly, approximately 11 times higher than that of ${\rm Ti}{{{\rm O}}_{2{\rm ,Co}}}$. This further confirms the success of the transformation of spherical ${\rm Ti}{{{\rm O}}_{2}}$ to nanotube ${\rm Ti}{{{\rm O}}_{2}}$ by using a concentrated NaOH solution.

Table 1. Isotherm adsorption curve and BET specific surface of ${\rm Ti}{{{\rm O}}_{{\rm 2}}}$ products.

${\rm Ti}{{{\rm O}}_{2}}$	Isotherm adsorption	BET surface (${{{\rm m}}^{2}}\,{{{\rm g}}^{-1}}$)
${\rm Ti}{{{\rm O}}_{2{\rm ,Co}}}$		9.6
TNT		105.9
BTA/TNT (norm.imp)		59.6
BTA/TNT (rota.eva)		25.2

In addition, it was found that the specific surface of BTA/TNT (norm.imp) is much smaller than that of TNT. This may be explained by the fact that BTA molecules entered the nanotube and occupied the tube space. Thus the specific surface of TNT precursor reduced by half. And BTA/TNT (rota.eva) specific surface has a quarter of TNT. This shows that the accumulation and lumping of BTA at TNT orifice have reduced significantly the free surface of TNT.

Maximum adsorption volumes of nitrogen on the adsorption and desorption isotherm curves illustrate further the specific surface area of each type of TNT.

The thermal analysis of BTA precursor is presentted in figure 4. On the DSC curve, a clear endothermic peak starts at $94{{\,}^{\circ }}{\rm C}$ and ends at $108{{\,}^{\circ }}{\rm C}$, while on the TGA curve, the BTA weight does not change within this temperature range. This indicates that $94{{\,}^{\circ }}{\rm C}$ is the melting point of BTA and the BTA is melted completely at $108{{\,}^{\circ }}{\rm C}$. Therefore, during the preparation, BTA/TNT products were only dried at a maximum temperature of $80{{\,}^{\circ }}{\rm C}$. The TGA curve also shows that BTA begins to decompose at $129{{\,}^{\circ }}{\rm C}$ and decomposes almost completely at $265{{\,}^{\circ }}{\rm C}$. The strongest thermal decomposition happens at $260{{\,}^{\circ }}{\rm C}$ with a mass loss rate of 24.1% min⁻¹ as on the derivative weight curve.

Zoom In Zoom Out Reset image size

TGA curves of TNT and BTA/TNT are displayed in figure 5. The TNT weight curve (figure 5(a)) indicates that the weight loss (ΔW) of 3.1% is due to the moisture evaporation, which occurs mainly within a temperature range of less than $100{{\,}^{\circ }}{\rm C}$. For BTA/TNT products, the weight loss is due to both steps: moisture evaporation and BTA decomposition. At $600{{\,}^{\circ }}{\rm C}$, BTA is decomposed totally. It is assumed that the amount of moisture in BTA/TNT products is primarily from TNT because of its hydrophilicity. BTA amount in BTA/TNT products is then calculated as follows:

$$\begin{align} \newcommand{\e}{{\rm e}} \displaystyle \begin{array}{@{}cccccccccccccccccccc@{}} \%{\rm BTA}=\%{\rm weight}\,{\rm loss}-\%\,{\rm moisture}\,{\rm evaporation}\quad \quad \,\, \nonumber \\ \%{\rm BTA}=\%\,{\rm weight}\,{\rm loss}-3.1\%*\%{\rm TNT}\quad \quad \quad \quad \quad \quad \nonumber \\ \%{\rm BTA}=\%\,{\rm weight}\,{\rm loss}-3.1\%*\left(100-\%\,{\rm weight}\,{\rm loss} \right) \nonumber \end{array}.\nonumber \end{align}$$

Zoom In Zoom Out Reset image size

The calculation results of BTA amount in BTA/TNT are shown in table 2. It was found that concentrations of BTA present in BTA/TNT products from vacuum impregnation at cooling temperature and ambient temperature are approximately the same, about 8%. Meanwhile, the BTA amount contained in BTA/TNT (rota.eva) is nearly 67%.

Table 2. Weight loss of TNT and BTA/TNTs.

${\rm Ti}{{{\rm O}}_{{\rm 2}}}$	% weight loss from $30{{\,}^{\circ }}{\rm C}$ to $600{{\,}^{\circ }}{\rm C}$	% weight of BTA in BTA/TNT products	% weight loss from $265{{\,}^{\circ }}{\rm C}$ to $600{{\,}^{\circ }}{\rm C}$
TNT	3.1	0	0.05
BTA/TNT (cool.imp)	10.8	8.0	7.7
BTA/TNT (norm.imp)	11.0	8.2	7.9
BTA/TNT (rota.eva)	67.6	66.6	4.5

The TGA curves of BTA/TNT (cool.imp) (figure 5(b)) and BTA/TNT (norm.imp) (figure 5(c)) demonstrate that the weight loss at maximum rate of about 0.3÷0.4% min⁻¹, occurs at approximately $510{{\,}^{\circ }}{\rm C}$. Meanwhile, BTA is almost completely decomposed at $265{{\,}^{\circ }}{\rm C}$ (figure 4). On the TGA curve of TNT, the mass loss at $510{{\,}^{\circ }}{\rm C}$ is very small, considered zero. Therefore, the loss at $510{{\,}^{\circ }}{\rm C}$ of vacuum impregnation products is undoubtedly due to the decomposition of BTA. This allowed us to conclude that, BTA is stored in the ${\rm Ti}{{{\rm O}}_{{\rm 2}}}$ tubes and is protected by TNT. As a result, the decomposition of BTA becomes more difficult and occurs at higher temperatures than pure BTA.

In figure 5(d), profiles of both the weight curve and the derivative weight curve of BTA/TNT (rota.eva) are very similar to that of the pure BTA in figure 4. A major loss occurs from $144{{\,}^{\circ }}{\rm C}$ to $265{{\,}^{\circ }}{\rm C}$ with a big value of 63.1%. In contrast, in table 2, the mass loss in $265\div 600{{\,}^{\circ }}{\rm C}$ range of BTA/TNT (rota.eva) is only 4.5%, while those of BTA/TNT (cool.imp) and BTA/TNT (norm.imp) are 7.7% and 7.9%, respectively. This result evidences that the BTA amount in BTA/TNT (rota.eva) is mainly outside the ${\rm Ti}{{{\rm O}}_{{\rm 2}}}$ tube but less stored in the tube.

Therefore, it can be concluded that almost all of BTA amount in BTA/TNT products obtained via the two impregnation methods is only decomposed at a temperature range of $265\div 600{{\,}^{\circ }}{\rm C}$, thanks to the protection of ${\rm Ti}{{{\rm O}}_{{\rm 2}}}$.

FT-IR spectrum of BTA in figure 6(a) with strong signals at $1209\,{\rm c}{{{\rm m}}^{-1}}$ and $752\,{\rm c}{{{\rm m}}^{-1}}$ shows triazole ring stretching and adjacent C-H wag deformation, respectively [15]. In figure 6(b), TNT's FT-IR spectrum gives the strongest signal at $790\,{\rm c}{{{\rm m}}^{-1}}$ indicating the stretching of Ti-O-Ti bond. In addition, signals at $3438\,{\rm c}{{{\rm m}}^{-1}}$ and $1621\,{\rm c}{{{\rm m}}^{-1}}$ show the valence and deformation vibration of O-H bond.

Zoom In Zoom Out Reset image size

For BTA/TNT products, all FT-IR spectra exhibit individual peaks of BTA and TNT. In particular, on the BTA/TNT (rota.eva) spectrum, BTA signals appear very intense, equivalent to BTA precursor. This shows that BTA/TNT (rota.eva) contains large amounts of BTA but BTA is mainly outside the TNT tube, so its signal is very clear.

For BTA/TNT (norm.imp) and BTA/TNT (cool.imp) products, the BTA signal is quite clear, but less intense because BTA concentrations are much smaller than TNT. In addition, BTA is stored in ${\rm Ti}{{{\rm O}}_{{\rm 2}}}$ tubes, so vibration signals of BTA bonds are weak because of the TNT obstacle.