Grafting of an aluminium surface with organic layers

Abstract

The grafting of organic films on an aluminum surface with its native oxide is demonstrated in protic or aprotic media by various methods: (i) spontaneous reduction of aryldiazonium salts, (ii) simultaneous electrochemical grafting of diazonium salts and alkyl iodides, (iii) spontaneous reaction of perfluoroalkylamine in an organic solvent and (iv) photochemical grafting of acetonitrile. The films are characterized by IRRAS, XPS, electrochemistry, water contact angles and ToF-SIMS; the latter demonstrates that aryl groups derived from diazonium salts are attached by an O–aryl bond. With all methods, the organic layers are strongly attached to the aluminum surface as they resist ultrasonic cleaning. Superhydrophobic films can be obtained when perfluoroalkyl groups are bonded on the surface.

---

Introduction

Aluminum is an important and widely used engineering metal that is protected by a native oxide layer (2–10 nm thick).^1–4 The functionalization of its surface has gained increased interest in a variety of applications. For example, anodising,⁵ powder coating or wet coating permit to enhance different properties of aluminum such as its visual appearance, its excellent corrosion resistance and its wear resistance. Different classes of organic molecules allow its surface modification such as fatty acids,^6,7 phosphonic acids^8–11 and silanes.^12,13 The spontaneous grafting of aluminum plates^14,15 and nanoparticles¹⁶ by diazonium salts in acetonitrile has also been reported. The stability of the organic film on the aluminum surface depends strongly on the interaction between the oxidized metallic surface terminated by hydroxyl groups^17–19 and the organic moieties.

In the case of fatty acids, there is an electrostatic interaction between the charged metal surface and anions derived from the organic molecules. In this way, a roughened and porous aluminum surface can be modified with the hydrophobic C₁₈ alkyl chains from stearic acid to give a super hydrophobic surface.²⁰

With phosphonic acids and silanes, chemical bonds are formed between surface O atoms and P or Si atoms, respectively. When the organic film attached on the aluminum surface contains reactive groups, further chemical modifications are possible. For example, an Al surface modified with [12-(4-benzophenone)-dodecyl]phosphonic acid,¹⁰ can be further transformed with different polymers, by reaction of the attached benzophenone moieties, under UV irradiation, with the spin-cast polymer film. In this way, it is possible to tune the hydrophobicity of the Al surface by choosing the appropriate polymer.¹⁰ A simple procedure to obtain highly hydrophobic surfaces is achieved in one step by dipping an aluminum substrate in a NaOH solution of a fluorinated silane.¹² However, a drawback of these methods is the possible hydrolysis of Si–O and P–O bonds in a basic medium.

The films formed during electrografting or spontaneous grafting of aryldiazonium salts on metals evidence a high stability due to the formation of a covalent bond between a C atom of the aromatic moiety and the metal surface.^21–23 These grafting reactions can be performed in an aprotic medium (for example acetonitrile –ACN–) and also in aqueous acidic and basic solutions.^24–26 In addition to carbon and metals, diazonium salts have also been reacted with oxide surfaces: ITO,^27–30 TiO₂,^31–33 SnO₂,^34,35 SiO₂,³⁶ MnO₂,³⁷ CuO¹⁴ and Fe₂O₃ ³⁸ to give metal oxide–(Ar)_n surfaces. Aluminum nanoparticles have also been modified by spontaneous reaction with alkyl iodides in toluene³⁹ and with diazonium salts in ACN; in the latter case both Al–aryl and Al–O–aryl bonds were detected by ToF-SIMS.¹⁶

More recently, it was shown on glassy carbon and gold that the reduction of an aryl diazonium ArN₂⁺ in the presence of an alkyliodide RI leads to the formation of (i) alkyl surfaces if the aryl radical cannot bind to the surface (because of steric hindrance) or, (ii) mixed aryl/alkyl surfaces metal–[Ar + R]. The reaction mechanism involves the formation of an aryl radical, abstraction of an iodine atom from the alkyl iodide to give an alkyl radical and grafting of both radicals on the surface.^40,41 This method permits to modulate the surface properties of metals by a proper choice of the two groups and of their relative concentrations.

Amines can be grafted on carbon and Pt surfaces by electrochemistry through an intermediate aminyl radical^42,43 to give metal–NHR surfaces, and on metallic oxides the same surface is obtained spontaneously through a nucleophilic substitution⁴⁴ of a metal–OH group.

ACN, a very common solvent, permits the modification of metallic surfaces either indirectly by electrochemical reduction of the 2,6-dimethylbenzenediazonium salt⁴⁵ or under UV irradiation.⁴⁶ In both cases, cyanomethyl radicals are the key-species that give organic films such as metal–[CH₂–CH(CN)]_n.

Herein, different modifications of the surface of aluminum are described: (i) spontaneous reaction of diazonium salts in acidic or basic aqueous media and also in ACN, (ii) electrochemical reduction of diazonium salts in the presence of an iodo alkane in ACN, (iii) spontaneous reaction of amines in ACN and, (iv) photochemical activation of ACN. Different types of reactions are involved: electron transfer, nucleophilic substitution and photochemical homolytic cleavage. The grafted layers present different surface functional groups that permit to adjust their surface chemistry and their hydrophilicity/phobicity. They are characterized by IRRAS, XPS, water angle measurements and ToF-SIMS.

The chemicals used to modify the aluminum surface are diazonium salts (4-nitro, 3-methyl, 3,5-bis trifluoromethyl and 4-carboxybenzenediazonium tetrafluoroborate salts), along with the iodoalkane and the amine and acetonitrile; their structure is presented in Scheme 1.

Scheme 1 Starting compounds.

Experimental

Chemicals

ACN Chromasolv (99.9%) was from Across Organics; acetone, 4-nitrobenzenediazonium tetrafluoroborate (DNO₂), hydrochloric acid (37%) and sodium hydroxide were from Sigma-Aldrich. Milli-Q water (>18 MΩ) was used for rinsing the plates and for the preparation of solutions. 1-Iodo-1H,1H,2H,2H-perfluorooctane (ICH₂CH₂C₆F₁₃, R_FI) 97% was purchased from Lancaster. 4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,11-Heptadecafluoroundecylamine (H₂N(CH₂)₃C₈F₁₇, R_FNH₂) 97% was purchased from Aldrich. 4-Carboxybenzenediazonium (DCOOH) and 3,5-(bis)-trifluoromethylbenzenediazonium (D(CF₃)₂) were prepared by standard procedures.¹⁰

Aluminum samples

Massive industrial aluminum discs (1 cm diameter) were polished using 0.04 μm alumina slurry on DP-NAP polishing cloth with a Mecatech 234 polishing machine (Presi). After the polishing step, they were rinsed with Milli-Q water, sonicated for 10 min in acetone, and dried under a stream of nitrogen. They were stored under ambient conditions before their modification. In some experiments, they were etched for 5 min in 0.1 M NaOH to roughen the surface.

Electrografting

These experiments were performed with an EG&G 263A potentiostat/galvanostat and an Echem 4.30 version software. The reference electrode was Ag/AgCl and the counter electrode a platinum foil. Prior to the analysis, the modified discs were rinsed with copious amounts of water and sonicated in acetone for 6 or 8 min.

Photografting

Photochemical modifications were achieved via a wide spectrum UV irradiation mercury Pen Ray lamp UVPC-86079 (20 mA AC, 4400 μW cm⁻² at 254 nm for a distance of ∼2 cm) with emission at 184.9 (3%, ∼150 μW cm⁻²), 253.6 (100%), 312.5–313.1, 356.0–356.3, 407.7 and 435.8 nm. Prior to irradiation, a thin film of neat ACN was deposited onto the metallic discs that were placed 4 cm under the UV light source. After UV irradiation, the discs were rinsed with ACN, acetone and ultrasonicated for 10 min in acetone.

Spontaneous modification

To obtain a roughened surface, some Al discs were treated for 5 min in 0.1 M aqueous NaOH solution, rinsed with acetone for 8 min in an ultrasonic bath and left for 2 h in a 10 mM solution of the diazonium salt either in (i) a 0.01 M H₂SO₄ solution, (ii) a buffer solution at pH = 9.3 (Na₂CO₃/NaHCO₃) or, in (iii) ACN. In acidic solution, the surface becomes yellow, but this colour disappears upon sonication in ACN and DMF. The same procedure was used with a 15 mM R_FNH₂ solution in ACN. Prior to the analysis, the modified discs were rinsed with copious amounts of water and sonicated in acetone for 8 min.

IRRAS instrumentation

Spectra of modified aluminum discs were recorded using a purged (low CO₂, dry air) Jasco FTIR-6100 Fourier Transform Infra Red Spectrometer equipped with MCT (mercury–cadmium–telluride) detector. For each spectrum, 1000 scans were accumulated with a spectral resolution of 4 cm⁻¹. The background recorded before each spectrum was that of a clean substrate. In the case of the modification in acidic or basic media the backgrounds were registered after the same treatment.

XPS instrumentation

X-ray photoelectron spectra were recorded using a Thermo VG Scientific ESCALAB 250 system fitted with a micro-focused, monochromatic Al Kα X-ray source (1486.6 eV) and a magnetic lens, which increases the electron acceptance angle and hence the sensitivity. The pass energy was set at 100 and 40 eV for the survey and the narrow regions, respectively. The Avantage software, version 4.67, was used for digital acquisition and data processing. The spectra were calibrated against C1s set at 285 eV.

Water contact angles

They were measured with a Kruss DSA3 instrument. A drop (3 μL) of Milli-Q water was automatically deposited on the top of the test sample placed in a horizontal position on the instrument stage. At least five measurements were made for each sample. The values of the contact angles were calculated by the tangent method using DropShapeAnalysis software.

ToF-SIMS

The spectra were acquired using a TOF-SIMS V spectrometer (IONTOF GmbH). The analysis chamber was maintained at less than 5 × 10⁻⁷ Pa under operating conditions. The total primary ion flux was below 10¹² ions cm² to ensure static conditions. A pulsed 25 keV Bi⁺ primary ion source (Liquid Metal Ion Gun, LMIG) at a current of about 1 pA (high current bunched mode), rastered over a scan area of 500 × 500 μm, was used as the analysis beam.

Results and discussion

The reactions scrutinized here involve radicals in solution on Al surfaces or nucleophilic attack of amines. As well documented in the literature, the surface of Al is constituted of Al–OH groups and Al–O oxides,^17–19 which both should react with radicals while only Al–OH groups would react with amines.

Spontaneous modification with diazonium salts in aqueous acidic, basic and ACN solutions

In an aqueous acidic medium and in aprotic non-nucleophilic solvents, Ar–N₂⁺ cations are present,⁴⁷ but at neutral and basic pHs, equilibria between Ar–N₂⁺ and the diazohydroxide, Ar–N=NOH, or the diazoate, Ar–N=N–O⁻, are established; the pK_as for the reaction of water on various Ar–N₂⁺ have been determined.⁴⁸ The dediazonation of diazonium salts occurs either homolytically or heterolytically; the homolytic cleavage gives a radical that reacts with the surface.⁴⁹

Open circuit potential (ocp) in an acidic medium. In ACN + 0.1 M NBu₄BF₄, in the absence of diazonium salt, the ocp of an Al electrode is ∼−0.75 V/Ag/AgCl. In an aqueous acidic solution containing 0.01 M H₂SO₄ it is ∼−0.85 V/Ag/AgCl (Fig. 1); upon addition of DNO₂ until a final concentration of 10⁻² M is reached the ocp increases to ∼−0.71 V/(Ag/AgCl).⁵⁰ These ocp values indicate that even in the presence of its native oxide/hydroxide, aluminum is able to reduce DNO₂ (E_p = 0.2 V/Ag/AgCl under the same conditions on a glassy carbon electrode) both in ACN and in an aqueous acidic solution. This ocp values preclude recording the voltammogram of DNO₂ on an Al electrode but permits the spontaneous grafting in both solvents. The jump of the potential is attributed to the electron transfer from Al to DNO₂ (Fig. 1).^51,52

Fig. 1 Ocp of an Al electrode vs. immersion time in a 0.01 M H₂SO₄ aqueous solution, the potential jump is observed after addition of DNO₂ (10 mM final concentration).

IRRAS. Fig. 2A shows the IR spectrum of an aluminum disc left for 1 h in a 0.01 M H₂SO₄ aqueous solution containing 10 mM DNO₂. Two IRRAS bands are observed at 1530 and 1350 cm⁻¹, which are attributed to the asymmetric and symmetric stretching of the –NO₂ group (1540 and 1356 cm⁻¹ for DNO₂) altogether with a band at 1592 cm⁻¹ that corresponds to the aromatic group; they confirm that grafting occurs in the acidic aqueous medium. Similar spectra are obtained in a pH = 9.3 buffer and in ACN + 0.1 M NBu₄BF₄ (not shown). This indicates that the same film is obtained from the diazonium salt in an acidic medium or in ACN, and from a diazoate in a basic medium.²⁴ The spectrum at pH = 9.3 also indicates that the film is stable in a basic medium unlike that obtained from silanes or phosphonic acids. However, the absorbance of the 1350 cm⁻¹ band is approximatively 30% lower in the basic medium and 7% lower in ACN than in the acidic solution.

Fig. 2 IRRAS spectra of Al discs after spontaneous modification in 0.01 M H₂SO₄ for 1 h by (A) DNO₂, (B) D(CF₃)₂ and (C) DCOOH. Ultrasonic cleaning for 8 min in acetone.

The grafting of an aluminum surface was also performed with D(CF₃)₂ in an aqueous acidic medium. The IR spectrum of the modified surface (Fig. 2B) exhibits the medium intensity vibrations of the aromatic ring at 1559 cm⁻¹, a very strong band at 1275 cm⁻¹ and medium ones at 1374, 1194, 1148 cm⁻¹ due to stretching vibrations of the C–F bonds (by comparison: 1598, 1355, 1276, 1206, 1139 cm⁻¹ for D(CF₃)₂). These bands confirm the presence of 3,5-bis-trifluoromethylphenyl moieties on the Al surface.

An Al surface was also modified with 10 mM DCOOH (formed in situ from the corresponding amine) in an aqueous acidic medium. The IR spectrum of the modified surface (Fig. 2C) presents a broad band in the 3000–3500 cm⁻¹ region with a peak at 3250 cm⁻¹ attributed to –OH stretching (insert in Fig. 2C), a strong band at 1712 cm⁻¹ due to –C=O stretching and a band at 1605 cm⁻¹ assigned to the aromatic ring puckering (3000–3500, 1720, 1615 cm⁻¹ for DCOOH).

XPS. An unmodified Al sample presents the contributions of Al2p (71.8 eV, 1.3% and 74.5 eV, 25.2%, corresponding respectively to the metal and the oxide/hydroxide⁵³), C1s (284.8 eV, 24.9%, from adventitious contamination), N1s (400.0 eV, 0.6%, from adventitious contamination), O1s (531.9, 47.7%, deconvoluted with two components at 531.8 eV and 532.7 eV corresponding, respectively, to Al₂O₃ and Al–OH¹⁷), and F1s (685.5 eV, 0.3%, likely corresponding to AlF₃, 3H₂O reported at 686.3 eV⁵⁰). The XPS survey spectrum (Fig. 3) of an Al plate modified with 10 mM DNO₂ in 0.01 M H₂SO₄ for 1 h shows the presence of Al2p (74.6 eV, 33.1%), S2p (169.3 eV, 2.7%), C1s (284.7 eV, 22,6%), N1s (399.8 eV 0.9%; 401.8 eV, 0.6% and, 405.5 eV, 1%), O1s (532.0 eV, 35.6% with the same two components at 531.8 and 532.7 eV as above) and F1s (686.5 eV, 3.5%). Al2p can be deconvoluted with two contributions corresponding to the oxide/hydroxide and the metal as in the unmodified sample. The S2p signal corresponds to Al₂(SO₄)₃ (171.0 eV) due to the reaction of H₂SO₄ with the surface.⁵⁴ The N1s signal presents the expected contribution at 405.5 eV corresponding to the nitro group; the 399.8 and 401.8 eV (protonated amine) signals can be assigned to (i) adventitious contamination as found in the pristine sample, and (ii) reduction of the NO₂ group by the X-ray beam,⁵⁵ and (iii) the presence of azo bonds in the film as already observed by XPS and ToF-Sims.⁵⁶ The signal of F1s assigned to AlF₃, 3H₂O probably stems from the preparation procedure of Al, but we have no explanation for the increase after grafting as during this reaction, the sample has never been in contact with a source of fluorine. The observation of Al contributions on the modified surface indicates a rather thin film (<5–10 nm).

Fig. 3 XPS spectrum of an Al surface with its native oxide modified by dipping for 1 h in a 0.01 M H₂SO₄ + 10 mM DNO₂ aqueous solution. Insert: N1s contribution.

ToF-SIMS spectra. Two Al discs were modified by reaction with 10 mM DNO₂ in 0.01 M H₂SO₄. The first one was modified for 1 min to obtain a very thin layer in order to observe the bond between the film and the Al surface (Fig. 4A–D). The second one was treated for 1 h to obtain a thicker layer to observe the organic layer on the surface (Fig. 4A′, B′ and E).

Fig. 4 ToF-SIMS spectra of an Al disc maintained in a 0.01 M H₂SO₄ solution for (A–D) 1 min, (A′, B′ and E) 1 h, (a) in the absence and (b) in the presence of 10 mM DNO₂. Ultrasonic cleaning for 6 min in acetone. (A) NO₂⁻, (B) C₆H₄NO₂⁻, (C) O–C₆H₄NO₂⁻, (D) Al–O–C₆H₄NO₂⁻, (E) C₆H₃NO₂–C₆H₃NO₂⁻.

The first disc presents the signals of NO₂⁻, C₆H₄NO₂⁻ and also of O–C₆H₄NO₂⁻ and Al–O–C₆H₄NO₂⁻. The latter two fragments indicate that the organic film is bonded through an oxygen atom to the alumina surface. There is no signal that would indicate a direct bonding of the film to an Al atom, contrary to what was observed with Al nanoparticles where part of the aryl groups was directly bonded to Al.³⁸ The difference originates from the preparation conditions of the Al samples: nanoparticles were prepared and modified in an oxygen free glove box, while the discs were handled in air.

When the oxidized aluminum surface has been maintained for 1 h in the acidic solution of the diazonium salt, aside from the NO₂⁻ and C₆H₄NO₂⁻ fragments, the C₆H₃NO₂–C₆H₃NO₂⁻ peak corresponding to the formation of a multilayer is observed. Such fragments have already been observed after spontaneous or electrochemical modification of a carbon surface.⁵⁷ However, the coverage of the surface is only partial as fragments pertaining to surface oxides are still obtained.

Water contact angles. The water contact angle, Θ, of a freshly polished Al surface is 33 ± 3°. An Al disk was maintained for 5 min in 0.1 M NaOH to increase the rugosity of the surface (Θ = 37.8 ± 5°), rinsed with distilled water and left for 100 min in a 0.01 M H₂SO₄ solution + 5% ACN (to permit the solubilization of the diazonium salt) in the presence of 0.1 M D(CF₃)₂. The water contact angle reaches Θ = 97 ± 3°; by comparison perfluorodecanethiols on a poly(2-hydroxyethyl methacrylate-co-ethylene dimethacrylate) hydroxylated surface give water contact angles up to 130°.⁵⁸ This likely indicates that the film is not perfectly compact.

The coupling of aryl radicals derived from diazonium salts to metal oxide surfaces forming an O–Ar bond has already been observed^14,36,37 and the growth of the layer results, as previously proposed on metals, from the attack of aryl radicals on the first grafted aryl layer.⁵⁵ On the surface of oxidized Al (partly Al–OH and partly Al₂O₃), the coupling can take place either by H abstraction by the aryl radical or direct attack on the oxygen of Al₂O₃ (Scheme 2).

Scheme 2 Spontaneous grafting of diazonium salts on Al native oxide.

Electrochemical modification with diazonium salts in the presence of alkyl halides

On an Au surface in ACN + 0.1 M NBu₄BF₄, the reduction of R_FI takes place at a quite negative potential, E_p = −2.0 V/Ag/AgCl. Such a potential is at the negative limit of the activity domain of an Al electrode in this medium, which precludes recording a voltammogram of this species. We have previously shown that it is possible to reduce alkyl halides with a large electrocatalytic effect reaching up to 2.1 V and simultaneously attach alkyl groups to carbon and metal surfaces. This has been achieved by reducing diazonium salts at potentials between +0.2 and −0.6 V per SCE, forming an aryl radical that abstracts an iodine atom and furnishes an alkyl radical that finally reacts with the surface.^39,40 Mixed surfaces were then formed by grafting at the same time an aryl radical from the diazonium salt and an alkyl radical from the alkyl halide to the surface and the first grafted layer.

Following the same process, DNO₂ (40 mM) was reduced electrochemically in the presence of R_FI (200 mM) by chronoamperometry at E = −0.8 V/Ag/AgCl for 30 min in an ACN + 0.1 M NBu₄BF₄ solution. At this potential, only the diazonium salt is reduced. However, the IRRAS spectrum of the modified Al disc (Fig. 5a) presents the signature of the perfluoroalkyl groups at 1246, 1212 and 1147 cm⁻¹ (1236, 1206, 1144 cm⁻¹ for I(CH₂)₂C₆F₁₃) and of the 4-nitrophenyl group at 1351 and 1529 cm⁻¹ (symmetric and asymmetric NO₂ stretching) and 1592 cm⁻¹ (ring vibration). With DCH₃, the peaks of the perfluoroalkyl groups are located at the same wavenumbers as above (Fig. 5b). The water contact angle of the film obtained from DNO₂ in the presence of I(CH₂)₂C₆F₁₃ is Θ = 116 ± 3°, indicating a highly hydrophobic film.

Fig. 5 IRRAS spectra of standard and roughened electrografted Al discs. (A) Al discs submitted to standard preparation. (B) Al disc roughened by basic treatment. Electrografting in the presence of I(CH₂)₂C₆F₁₃ and (a) DNO₂, (b) DCH₃, (c and d) D(CF₃)₂ (c) before and (d) after 4 h extraction in boiling toluene. Ultrasonic cleaning for 6 min in acetone.

To increase the rugosity of the Al surface prior to any functionalization reaction, an Al disc was treated with 0.1 M NaOH for 5 min. Then, D(CF₃)₂ (40 mM) was reduced in the presence of R_FI (200 mM) in ACN by chronoamperometry at E = −0.8 V per SCE for 15 min on such an Al disc. The IRRAS spectrum of the modified surface (Fig. 5c) presents the signal of aryl–CF₃ group at 1282 cm⁻¹ and the perfluoroalkyl bands at 1246, 1212 and 1147 cm⁻¹. The XPS survey spectrum (Fig. ESI. 1†) shows the contributions of Al2p (74.9 eV, 20.6%), C1s (285.0 eV, 25.0%), N1s (400.2 eV, 1.0%), O1s (532.2 eV, 46.8%) and F1s (685.6 eV, 6.6%). The Al2p peak, which corresponds to the oxidized form only, can be deconvoluted into two contributions: 74.9 eV (17.8%) for Al₂O₃ + Al–OH^17,18,51 and 75.6 eV (2.33%) for AlF₃ (75.80 eV (ref. 50)). C1s presents a number of contributions in agreement with the complex structure of such films⁴⁰ among which CF₃ (293.6 eV) is clearly identified by comparison with the XPS spectrum of R_FI grafted on GC (293.7 eV³⁹) and that of trifluoromethylbenzene (293.8 eV⁵⁰). F1s can be deconvoluted in two peaks at 685.6 (5.2%) for AlF₃, 3H₂O and 688.8 eV (1.2%) for the perfluoroalkyl group (689.7 eV for polytetrafluoroethylene⁵⁹ and 690.8 eV (ref. 60) for C₆H₅CF₃). In order to find out if the presence of AlF₃, 3H₂O could be due to the reaction of the perfluoroalkyl compound with aluminum, perfluorooctane was deposited on the surface of Al and left for 30 min; after rinsing, XPS does not show any increase of the AlF₃ contribution (not shown).

Using both perfluorinated alkyl and trifluoromethyphenyl groups, the surface becomes superhydrophobic with a water contact angle Θ = 161 ± 3° (Fig. ESI. 2†).

The stability of these films was tested by washing out the modified roughened plate for 4 h with toluene in a Soxhlet. The Al disc modified film by the perfluoroalkyl chains and the 3,5-bis trifluoromethylphenyl groups only shows a small decrease of the IRRAS intensity (less than 5% at 1246 cm⁻¹, Fig. 5d), but the spectrum is better defined than that before washing, indicating the removal of some adsorbed species during the extraction with boiling toluene.

Upon one electron reduction on glassy carbon, the diazonium salt is reduced to an aryl radical that abstracts a iodine atom from RI to give a R˙ radical.³⁹ This permits to reduce the alkyl iodide with a huge electrocatalytic effect (∼2.1 V in the case of DNO₂). On the Al discs, the potential was maintained at −0.8 V/Ag/AgCl to ensure the reduction of the diazonium salt to a radical prone to iodine atom abstraction from the alkyl iodide. Therefore, both aryl and alkyl radicals should be obtained as on glassy carbon and both attach to the Al surface. The structure of the obtained multilayers should be analogous to that already observed on glassy carbon (Scheme 3).³⁹

Scheme 3 Electrografting of diazonium salt and alkyl iodide on Al native oxide.

Spontaneous modification with amines

The two spontaneous grafting reactions reported above rely on the reduction of diazonium salts by an electron transfer from Al through the surface oxide film. We now report a surface modification that is not based on an electron transfer but on a nucleophilic substitution of hydroxyl groups. An Al disc treated previously with a 0.1 M NaOH aqueous solution for 5 min to increase the rugosity of the surface was thereafter immersed for 2 h into a solution of ACN containing 15 mM R_FNH₂. The IRRAS spectrum of the modified Al disc (Fig. 6) presents the NH deformation peak at 1583 cm⁻¹ and the CF stretching peaks at 1248, 1217 and 1152 cm⁻¹.⁶¹ The XPS survey spectrum (Fig. ESI. 3†) shows the contribution of Al2p (74.7 eV, 23.4%), C1s (284.9–293.6 eV), N1s (400.2 eV, 1.3%), O1s (532.3 eV, 34.9%) and F1s (685.6 eV, 21.6%). Al2p can be deconvoluted in three peaks at 71.6 (1.75%), 74.6 (16.0%) and 75.4 eV (5.4%) corresponding respectively to Al, Al₂O₃ + Al–OH and AlF₃ as above. The C1s peak can be deconvoluted with a peak at 293.6 eV corresponding to the terminal CF₃.⁴⁰ The 400.2 eV N1s peak is close to the 398.8 eV reported value for the N–metal bond on iron and copper.⁴³ F1s can be deconvoluted in two peaks at 685.4 eV (16.0%) and 688.8 eV (5.6%) corresponding respectively to AlF₃, 3H₂O as above and the perfluoroalkyl group. The water contact angle is Θ = 110 ± 2°.

Fig. 6 IRRAS spectrum for an Al disc treated in an aqueous basic solution and immersed for 2 h into an ACN + 15 mM H₂NC₂H₄C₈F₁₇ solution. Ultrasonic cleaning for 8 min in acetone.

The attack of amines on copper or iron surface oxides is reported to involve the nucleophilic substitution of a M–OH group (M = metal) by RNH₂ to give M–NHR.⁴³ We assume that the same mechanism takes place on Al where OH groups constitute a high fraction of the surface.¹⁷ In organic chemistry, unprotonated hydroxyl are poor leaving groups, but Al is much more electropositive than carbon (electronegativity 1.63 vs. 2.55), which increases the polarity of the bond and favors the cleavage and substitution of the OH group (Scheme 4).

Scheme 4 Spontaneous grafting of amines on Al native oxide.

Photochemical modification with acetonitrile

The IRRAS spectrum of an Al disc immersed in ACN and irradiated with UV light for 45 min (Fig. 7a) presents the same features as those obtained when gold, copper, nickel and iron samples are modified under photochemical irradiation of ACN.⁴⁵ The two strong bands at ∼3250 and ∼1630 cm⁻¹ correspond to –NH₂ stretching and –NH₂ deformation respectively and the band at ∼1400 cm⁻¹ is attributed to –CH₂ scissoring vibrations. These bands are not present when the Al disc has been maintained for the same time in ACN without irradiation (Fig. 7b). This IR spectrum indicates the modification of the surface by amino groups.

Fig. 7 IRRAS spectra of Al discs immersed in ACN for 45 min (a) under irradiation or (b) without irradiation. Ultrasonic cleaning for 8 min in acetone.

The reaction is triggered by the very reactive cyanomethyl radical, ˙CH₂CN obtained by irradiation of ACN (that strongly absorbs below 190 nm). This radical abstracts a hydrogen atom from the hydroxylated surface; the coupling of the oxygen and cyanomethyl radicals provides a bonded Al–O–CH₂CN group. The growth of the layer can follow two different paths involving either the ˙CH₂CN radical or the –CH₂CN anion^44,45 that should be produced by reduction of ˙CH₂CN. Because of the values of the radical/carbanion couple (E°(˙CH₂CN/⁻CH₂CN) = −0.66 V/Ag/AgCl)⁶² and of the ocp of the Al native oxide surface in ACN (E_ocp(Al_ACN) = −0.75 V/Ag/AgCl) are close, both routes are possible (Scheme 5).

Scheme 5 Photografting of acetonitrile on Al native oxide.

Conclusion

Four different approaches are presented for the modification of the Al native oxide surface by (i) spontaneous grafting of aryl diazonium salts (prepared ex or in situ) in aqueous acidic or basic media and in ACN; (ii) electrografting of an alkyl iodide in the presence of a diazonium salt, leading to mixed alkyl/aryl films; (iii) spontaneous grafting of amines; (iv) photografting of acetonitrile that gives aminated surfaces. Owing to (ii), the proper choice of the two starting compounds permits to obtain superhydrophobic surfaces. Different mechanisms are involved: reaction of radicals on the oxidized Al surface (aryl, perfluoroalkyl and ˙CH₂CN radicals) or nucleophilic attack of Al by amines. The radicals are obtained either by spontaneous or electrochemical reduction of diazonium salts. The layers characterized by IRRAS and XPS are strongly attached to the Al surface as they resist sustained ultrasonic cleaning. The aryl groups (obtained from diazonium salts) are bonded to the oxidized Al surface via an oxygen atom, as evidenced by Al–O–Ar fragments in ToF-SIMS experiments. These methods are very simple and quite general; they use easily available starting compounds and should therefore find specific applications in the surface modification of Al.

References

1. J. R. Davis, Aluminum and Aluminum Alloys, ASM International, 1993.
2. E. Ghali, Corrosion resistance of aluminum and magnesium alloys, Wiley, Hoboken, 2010.
3. R. Strohmeier, Surf. Interface Anal., 1990, 15, 51.
4. M. R. Alexander, G. E. Thompson, X. Zhou, G. Beamson and N. Fairley, Surf. Interface Anal., 2002, 34, 485.
5. W. Lee and S.-J. Park, Chem. Rev., 2014, 114, 7487.
6. I. Liascukiene, N. Aissaoui, S. J. Asadauskas, J. Landoulsi and J.-F. Lambert, Langmuir, 2012, 28, 5116.
7. K. Nielsch, J. Choi, K. Schwirn, R. B. Wehrspohn and U. Gösele, Nano Lett., 2002, 2, 677.
8. M. J. Pellerite, T. D. Dunbar, L. D. Boardman and E. J. Wood, J. Phys. Chem. B, 2003, 107, 11726.
9. J. Pahnke and J. Rühe, Macromol. Rapid Commun., 2004, 25, 1396.
10. M. Giza, P. Thissen and G. Grundmeier, Langmuir, 2008, 24, 8688.
11. C. Schmitt Pauly, A.-C. Genix, J. G. Alauzun, M. Sztucki, J. Oberdisse and P. H. Mutin, Phys. Chem. Chem. Phys., 2015, 17, 19173.
12. N. Saleema, D. Sarkar, R. Paynter and X. Chen, ACS Appl. Mater. Interfaces, 2010, 2, 2500.
13. N. Saleema, D. K. Sarkar, D. Gallant, R. W. Paynter and X.-G. Chen, ACS Appl. Mater. Interfaces, 2011, 3, 4775.
14. B. L. Hurley and R. L. McCreery, J. Electrochem. Soc., 2004, 151, B252.
15. A. M. Mahmoud, A. J. Bergren and R. L. McCreery, Anal. Chem., 2009, 81, 6972.
16. Y. Aït Atmane, L. Sicard, A. Lamouri, J. Pinson, M. Sicard, C. Masson, S. Nowak, P. Decorse, J.-Y. Piquemal, A. Galtayries and C. Mangeney, J. Phys. Chem. C, 2013, 117, 26000.
17. E. McCaferty and J. P. Wightman, Surf. Interface Anal., 1998, 26, 549.
18. J. van den Brand, P. C. Snijders, W. G. Sloof, H. Terryn and J. H. W. de Wit, J. Phys. Chem. B, 2004, 108, 6017–6024.
19. Ö. Özkanat, B. Salgin, M. Rohwerder, J. M. C. Mol, J. H. W. de Wit and H. Terryn, J. Phys. Chem. C, 2012, 116, 1805.
20. L. Feng, H. Zhang, P. Mao, Y. Wang and Y. Ge, Appl. Surf. Sci., 2011, 257, 3959.
21. Aryl Diazonium Salts. New Coupling Agents in Polymer and Surface Science, ed. M. M. Chehimi, Wiley-VCH, Weinheim, 2012.
22. D. Bélanger and J. Pinson, Chem. Soc. Rev., 2011, 40, 3995.
23. J. Pinson and F. Podvorica, Chem. Soc. Rev., 2005, 34, 329.
24. C. Combellas, M. Delamar, F. Kanoufi, J. Pinson and F. I. Podvorica, Chem. Mater., 2005, 17, 3968.
25. F. I. Podvorica, F. Kanoufi, J. Pinson and C. Combellas, Electrochim. Acta, 2009, 54, 2164.
26. A. Berisha, F. Kanoufi, J. Pinson, C. Combellas and F. I. Podvorica, Electrochim. Acta, 2011, 56, 10762.
27. S. Maldonado, T. J. Smith, R. D. Williams, S. Morin, E. Barton and K. J. Stevenson, Langmuir, 2006, 22, 2884.
28. A.-M. Haque and J. K. Kim, Langmuir, 2011, 27, 882.
29. A.-M. Haque and J. K. Kim, Chem. Commun., 2011, 47, 6855.
30. F. Mirkhalaf, T. J. Mason, D. J. Morgan and V. Saez, Langmuir, 2011, 27, 1853.
31. A. Merson, T. Dittrich, Y. Zidon, J. Rappich and Y. Shapira, Appl. Phys. Lett., 2004, 85, 1075.
32. A. Mesnage, M. Abdel Magied, P. Simon, N. Herlin-Boime, P. Jégou, G. Deniau and S. Palacin, J. Mater. Sci., 2011, 46, 6332.
33. D. Quinton, A. Galtayries, F. Prima and S. Griveau, Surf. Coat. Technol., 2012, 206, 2302.
34. A. Adenier, N. Barré, E. Cabet-Deliry, A. Chaussé, S. Griveau, F. Mercier, J. Pinson and C. Vautrin-Ul, Surf. Sci., 2006, 600, 4801.
35. F. Lamberti, S. Agnoli, L. Brigo, G. Granozzi, M. Giomo and N. Elvassore, ACS Appl. Mater. Interfaces, 2014, 6, 22769.
36. C. Bureau and J. Pinson, Procédé de modification de surfaces isolantes ou semi-conductrices, et produits tels qu'obtenus, European patent, EP1948720, 2006.
37. K. J. Bell, P. A. Brooksby, M. I. J. Polson and A. J. Downard, Chem. Commun., 2014, 50, 13687.
38. K. Brymor, J. Fouineau, A. Eddarir, F. Chau, N. Yaacoub, J.-M. Grenèche, J. Pinson, S. Ammar and F. Calvayrac, J. Nanopart. Res., 2015, 17, 438.
39. M. Fogliazza, L. Sicard, P. Decorse, A. Chevillot-Biraud, C. Mangeney and J. Pinson, Langmuir, 2015, 31, 6092.
40. D. Hetemi, F. Kanoufi, C. Combellas, J. Pinson and F. I. Podvorica, Langmuir, 2014, 30, 13907.
41. D. Hetemi, H. Hazimeh, P. Decorse, A. Galtayries, C. Combellas, F. Kanoufi, J. Pinson and F. I. Podvorica, Langmuir, 2015, 31, 5406.
42. B. Barbier, J. Pinson, G. Desarmot and M. Sanchez, J. Electrochem. Soc., 1990, 101, 1757.
43. A. Adenier, M. M. Chehimi, I. Gallardo, J. Pinson and N. Vilà, Langmuir, 2004, 20, 8243.
44. I. Gallardo, J. Pinson and N. Vilà, J. Phys. Chem. B, 2006, 110, 19521.
45. A. Berisha, C. Combellas, F. Kanoufi, J. Pinson, S. Ustaze and F. I. Podvorica, Chem. Mater., 2010, 22, 2962.
46. A. Berisha, C. Combellas, G. Hallais, F. Kanoufi, J. Pinson and F. I. Podvorica, Chem. Mater., 2011, 23, 3449.
47. C. Galli, Chem. Rev., 1988, 88, 765.
48. A. Sienkiewicz, M. Szymula, J. Narkiewicz-Michaleka and J. Bravo-Díaz, J. Phys. Org. Chem., 2014, 27, 284.
49. R. Pazo-Llorente, H. Maskill, C. Bravo-Díaz and E. Gonzalez-Romero, Eur. J. Org. Chem., 2006, 2201.
50. D. Mercier, M. Herinx and M. G. Barthés-Labrousse, Corros. Sci., 2010, 52, 3406.
51. F. Le Floch, J.-P. Simonato and G. Bidan, Electrochim. Acta, 2009, 54, 3078.
52. J. Lehr, B. E. Williamson, B. S. Flavel and A. J. Downard, Langmuir, 2009, 25, 13503.
53. NIST X-ray Photoelectron Spectroscopy Database. NIST Standard Reference Database 20, Version 4.1, data compiled and evaluated by A. V. Naumkin, A. Kraut-Vass, S. W. Gaarenstroom and C. J. Powell.
54. J. A. Treverton and N. C. Davies, Electrochim. Acta, 1980, 25, 1571.
55. A. Adenier, E. Cabet-Deliry, A. Chaussé, S. Griveau, F. Mercier, J. Pinson and C. Vautrin-Ul, Grafting of Nitrophenyl Groups on Carbon and Metallic Surfaces Without Electrochemical Induction, Chem. Mater., 2005, 17, 491–501.
56. P. Doppelt, G. Hallais, J. Pinson, F. Podvorica and S. Verneyre, Chem. Mater., 2007, 19, 4570–4575.
57. C. Combellas, F. Kanoufi, J. Pinson and F. I. Podvorica, Langmuir, 2005, 21, 280.
58. L. Li, J. Li, X. Du, A. Welle, M. Grunze, O. Trapp and P. A. Levkin, Angew. Chem., Int. Ed., 2014, 53, 3835.
59. High resolution XPS of organic polymers. The Scienta ESCA300 Database, ed. G. Beamson and D. Briggs, John Wiley, Chichester, UK, 1992.
60. D. T. Clark, D. Kilcast and W. K. R. Musgrave, J. Chem. Soc. D, 1971, 516.
61. G. Socrates, Infrared and Raman Characteristic Group Frequencies, John Wiley, New York, 3rd edn, 2001.
62. A. A. Isse and A. Gennaro, J. Phys. Chem. A, 2004, 108, 4180.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra15313e

---