Layered double hydroxides (LDHs) as functional materials for the corrosion protection of aluminum alloys: A review

The search for new alternatives to replace chromate-based coatings is a matter of great importance. Since their oﬃcial ban due to the raised concerns of hexavalent chromium to the human health and the environment, signiﬁcant effort s have been devoted to ﬁnding a more suitable alternative for the corrosion protection of aluminum alloys. However, the task has been quite challenging since the potential replacement needs to fulﬁll several requirements both in terms of cost and exceptional corrosion performance. Layered double hydroxides (LDHs) have generated a lot of interest in the past few years. They have been proposed as prospective candidates to replace chromate based protective systems. The particular struc- ture of LDH nanocontainers allows them to intercalate a number of corrosion inhibitors and release them on demand under the action of corrosion relevant triggers. Moreover, their ﬂexible use as pigments in paints or as a pre-treatment directly as conversion layers makes their implementation even more con-venient. This review presents a critical view to the studies performed till today on LDHs for corrosion protection of aluminum alloys.

The ability of LDHs to be used in their derived form or/and as incipient (host for different species) along with their other interesting properties make them valuable for an extended range of functionalities [ 57 , 58 ]. Some of the interesting LDH properties can be listed as follows: (1) LDHs biocompatibility, prolonged buffering effect, antacids and ability to host pharmaceutically active molecules (e.g. diclofenac sodium, gemfibrozil, ibuprofen, naproxen, tolfenamic) were extensively investigated and reported in the area of drug delivery [58][59][60][61][62] . (2) The high surface area, homogeneous and thermally stable composition of the metal cations of the LDHs, makes them good candidates as catalysts [63][64][65] . (3) LDHs highly tunable architecture, non-toxicity and anion exchange properties are exploited as adsorbent for inorganic contaminants (eg. AsO 4 3 − , CrO 4 2 − , F − , Br − , I − ) in water purification [ 58 , 66 , 67 ]. (4) The crystallinity, morphology and conductivity (depending on the LDH composition) of LDH contribute to enhance the electrochemical performance of materials for energy conversion and storage in the form of electrodes for batteries, fuel cells, super capacitors and so on [68][69][70][71][72] .
This review focuses on the use of LDH for corrosion protection of aluminum alloys. Because of the ban of toxic chromatebased corrosion protective systems [73] , there is a growing need to redirect the research in the field of corrosion protection into more environmentally friendly concepts. Moreover, these systems should be cost-effective while maintaining a good performance in terms of corrosion protection, hence the recent developments in the area of self-healing "smart" approaches. The latter include the use of nanocontainers [ 51 , 74-78 ] that have the aptitude to host and release healing agents in a controlled manner. Due to the above listed properties and the remarkable anion-exchange capacity, LDHs have been dominating the area of active corrosion protection for a few years now [ 8 , 9 , 13 , 34 , 48 , 79-81 ]. The protective action of LDH loaded with inhibitive anions is based on the anion-exchange reaction induced by particular triggers, such as pH [ 30 , 82 ] and/or presence of aggressive species [ 35 , 83 ]. The operative mode of LDH nanocontainers involve (a) the release of inhibiting anions [ 53 , 84 ] while (b) capturing corrosive species such as Cl − [ 15 , 83 , 85 ] and/or (c) cathodic formation of a hydroxide film [86] ( Fig. 1 ).
This review begins with an overview of the former Cr(VI)based protective systems and their mechanism of action in Al alloy. A short and comprehensive outline in respect to the corrosion protective strategies proposed as potential replacement to Cr(VI), will be given next. Finally, a critical review of the studies accomplished on LDH as a "smart" active corrosion protective system for Al alloys is provided. The literature review of LDH will be divided in two main parts: first, discussing the studies on the application of LDH in form of pigment and second, on LDH applied as a conversion layer. At the end, a concise outlook on the overall application of LDH for corrosion protection of Al alloy and their current or/and possible implementation in industry, is addressed.

Cr (VI) based coatings
Chromate-containing coatings were extensively used in the past in many different applications. In the specific case of aerospace industry, they have perhaps been the main tool against corrosion for nearly 100 years [ 73 , 87 ]. These coatings are very effective, providing corrosion protection to critical aerospace parts (fuselage, wings, engines, etc.), that are required to operate in highly demanding environments for extended periods. The remarkable corrosion resistance of Cr-based systems is based on their so-called "self-healing" properties [88] . As the term "self-healing" can be defined in various ways, it is worth specifying that in the current context of corrosion, the term refers to the capacity of a coating system to heal (complete to partial recovery of the coating's main function, which is corrosion protection [ 89 , 90 ]) by releasing corrosion inhibiting species into the area of the defect or damage [90] .
Recent interpretations of the formation of chromate conversion coatings (CCC) on the surface of Al alloy suggest to be based on a sol-gel mechanism that involves three steps: hydrolysis, polymerization and condensation of Cr(III) [ 88 , 91 , 92 ]. Frankel et al. [88] explained that the formation of CCCs begins with the reduction of Cr(VI) to Cr(III) followed by a series of condensation reactions leading to the formation of a Cr(III) oxy-hydroxides polymer chain. The accumulation of high amounts of Cr(VI) on the Cr(OH) 3 polymer chain during the polymerization as a result of nucleophile attacks of OH − ligands, leads to the establishment of covalent Cr(VI)-O-Cr(III) bonds. In this way the Cr(VI) is stored and released when necessary ( Fig. 2 ) [ 88 , 91 ]. In most cases, additional anions (fluorides, phosphates, ferricyanides etc.) are added into the working baths to accelerate the formation of CCCs by preventing the Al surface passivation and promoting a faster reduction of the Cr(VI) to Cr(III) [87] .
The formation mechanism of CCC is strongly related with its functionality to provide active corrosion protection. Indeed, the  [88] ), followed by their mechanism of inhibition/self-healing on AA2024 upon mechanical damage [ 88 , 93 , 94 ].
previously mentioned reaction leading to the formation of the mixed oxides Cr(III) / Cr(VI) is reversible. Hence, upon contact with an aggressive media, the Cr(VI) is released and transported to the affected zones to inhibit any corrosion activity.
A similar operating mode is found for chromates stored in primers or/and topcoats (generally in form of SrCrO 4 ). Cr(VI) are leached in form of CrO 4 2 − anions into the electrolyte to undergo a corrosion inhibition process. An illustration of the different discussed mode of intervention of Cr-based systems with their different speciation is provided in Fig. 3 [95] . The first mode of action ( Fig. 3 a) is associated to the chromate capacity to function in wide range of pH, which in turn makes it a good candidate to be incorporated into different polymer matrices. When incorporated into a primer in form of SrCrO 4 ( Fig. 3 b) or used directly as a conversion coating CCC ( Fig. 3 c), chromates offer a self-healing functionality by the release of CrO 4 2 − into the surface upon mechanical damage or defect on the coating [95] .
The operating principle of Cr(VI) inhibition relies on a complex dual function mechanism. It has the aptitude to impede the rate of cathodic reduction reactions whilst mitigating the anodic metal dissolution ( Fig. 2 ) [ 92 , 96 ]. For instance, in high strength Al alloys, studies reported that the cathodic inhibition involves the reduction of Cr(VI) to Cr(III) which adsorbs on the Cu-rich particles and suppress the oxygen reduction reaction (ORR) [ 92 , 97 ]. The formed layer is irreversible, therefore even after removing chromates from the solution, the ORR stays inhibited [88] . When it comes to anodic inhibition, it is considered more to be a consequence of the cathodic inhibition process rather than actually an independent process. Hurley et al. [92] stated that the inhibition of the oxidation reactions related to the anodic dissolution of Al is a result of a general hindrance of the electron transfer imposed by the formation of a protective Cr (III) monolayer during the above mentioned cathodic inhibition process.
Despite the exceptional corrosion protective properties of chromate-based coatings, their use is highly restricted due to the striking toxicity for humans and the environment [98] . Cr(VI) was first listed as a human carcinogen in 1980 on the annual report published by the National Toxicology Program and the Depart- ment of Health and Human Services (US) [99] . The restrictions for the use of chromate in automotive and aerospace industries were being gradually issued through the following years. On the 21 st of September 2017, under the Europe's Registration, Evaluation, Authorization & Restriction of Chemicals (REACH) regulation, chrome-plating chemicals such as sodium dichromate were completely banned. The ban of some other chromate containing compounds such as strontium chromate used in primers followed on January 2019 [73] .
However, the replacement of a well-known technologically relevant additive such as chromate, with new alternatives is challenging especially for critical applications.
Therefore, the European Chemicals Agency (ECHA) has extended the sunset date for the use of Cr(VI) plating treatment for aerospace to the year 2024 and a few Cr(VI) additives for paints to the year 2026 [73] .
The performance of an alternative is expected to be comparable or even exceed the corrosion resistance of chromate-based systems and should comply with several requirements, some of which are: -Applicable in a wide range of pH (which makes it possible to be stored in primers and paints). -Fitting the minimal requirements for concentrations (in aqueous solution the concentration of Cr(VI) is generally between 0.05 to 0.075 M and are found in form of CrO 3 with a concentration of approx. 50 mM in coating formulations [91] ). -Similar or less time for processing steps in comparison to Crbased processes [ 100 , 101 ]. -Low bath temperatures 60 °C > T, ideally room temperature (RT) [ 100 , 101 ]. -Cost-effective and more importantly ecologically acceptable [95] . -Suitable for all relevant Al alloys [100] .
Over the course of many years, a large number of studies devoted to potential alternatives for chromates have been developed and a number of reviews [ 95 , 102 , 103 ] summarizing these alternatives were written. A list of a few examples of the most discussed alternatives that were applied to Al alloys is provided in the next section.

Alternatives to Cr (VI)-based coatings
Since the ban of Cr(VI)-based coatings, an important number of possible alternatives were proposed and some are already available commercially. In the following sections some examples are described and a summary is presented in Table 1 .

Trivalent chromium
Following the ban of Cr(VI), many studies focused on the use of Cr(III), considering it as the "second-best" option [104] . Trivalent chromium pre-treatments are commonly available as zirconium/trivalent chromium-based conversion coatings (TCC). The formation of the latter conversion coating takes place in an acidic bath (pH 3.8-4) generally containing trivalent chromium, zirconate, fluoride and sulfate salts. The mechanism of formation relies on a co-precipitation of hydrated zirconia and Cr(OH) 3 initiated by an increase of pH at the active cathodic sites (e.g. Cu-rich intermetallic in the case of AA2024) [105] .
Qi et al. [105] performed a deeper investigation on the composition of the Cr(III)-CC and its influence on the mechanism of corrosion protection of Al alloys. The authors found that the coating is composed of two layers: the thick outer layer consists of compounds such as AlF 3 Cr(OH) 3 , ZrO 2 and ZrF 4 (depending on the starting species on the bath) and the inner layer is mostly aluminium-rich and is responsible for providing the main corrosion protection which is mostly barrier-based.
Although, no Cr(VI) is used for the TCC growth process, some concerns were raised due to the high possibility of Cr(III) oxidizing to Cr(VI) [105][106][107] . For instance, using Raman microprobe spectroscopy method, Li et al. [107] revealed the presence of Cr(VI) oxide species formed via oxidation of Cr (III) in the presence of hydrogen peroxide. The authors linked the production of hydrogen peroxide to the local reduction of dissolved oxygen at the Cu-rich intermetallic. Qi et al. [108] also detected Cr(VI) species in an airaged coating after exposure to a corrosive electrolyte (NaCl) and it was associated to corrosion processes. However, Cr(III)-based treatments are still being extensively employed in industry [ 95 , 109 , 110 ], as the estimated amount of Cr(VI) found is minimal ( < 0.1 wt. % which complies with to REACH regulations [111] ),therefore presenting no risk. Moreover, the toxicological data published in the last few years do not allege any high risk to human health or the environment related to Cr(III). Nevertheless, some uncertainties are Table 1 Summary of possible alternatives to Cr(VI)-based coatings discussed in this section. Coat

3-6 min
NaF, KF, HF -Barrier protection, no self-healing effect [ 157 , 193 , 194 ] * Temp. and * acc. are short for temperature and accelerators, respectively still persistent and the non-self-healing aspect of the Cr(III)-systems could be a disadvantage when competing with other proposed alternatives to chromate.

Phosphates
Phosphate treatments were originally based on a mixture of chromate/phosphate conversion coatings that were known for their outstanding corrosion resistance and adhesion properties. Due to its banning, chromate was replaced by zinc, and the resulting zinc/phosphate coatings continued to be applied widely on steel [ 103 , 112 ] and were further extended to Al alloys [113][114][115][116][117] .
For the phosphating process, an acidic bath typically composed of phosphoric acid, zinc dihydrogen phosphate, fluoride and an oxidation accelerator (e.g. NO 2 ) is used [ 103 , 112 ].
Akhtar et al. [ 113 , 114 ] showed that the phosphating process on Al alloys is a two-step process which starts at the active sites by the dissolution of Al (1) and an increase of the local pH (2). This increase of pH will lead to a shift in the hydrolytic equilibrium between the soluble primary phosphates and the insoluble tertiary phosphates of the heavy metals, driving into the systematic conversion and deposition of the latter species and the generation of the phosphate coating (3) and (4) [ 113 , 114 ]: The final obtained Zn 3 (PO 4 ) 2 coating is reported to be reasonably hard, electrically non-conductive, adherent and with a good wear and corrosion resistance [112] . However, some studies have revealed a number of cons to the use of phosphate-based conversion coatings. Phosphate coatings generally present limited stability at higher pH and their destruction may lead to disbonding issues, if used under a polymer layer [ 118 , 119 ]. More critical, the waste resulting from the phosphating process is considered hazardous by the European Union as well as the US environmental protection agency. The phosphate coating sludges contain an important amount of heavy metals which makes the disposal of the phosphating baths detrimental [120] . With the presented evidence and the fact that phosphate coatings can offer only a barrier protection, phosphating may no longer be viable as an alternative to Cr(VI) based coatings.
It was suggested that the inhibition mechanism of cerium salts is a localized process that interferes with the corrosion reactions occurring around the intermetallic (IMC). In an investigation performed by Yasakau et al. [130] , it was shown that an increase of the local pH due to the cathodic process at the interface of AA2024, leads to the formation of cerium hydroxo-complexes. This continuous increase of pH until approx. 10 resulted in the formation of cerium hydroxide deposits on top of the S-phase IMCs, thus obstructing both the cathodic and anodic processes on the IMCs [130] . It was also pointed out that the advantage of cerium relies on its aptitude to be oxidized from Ce(III) to Ce(IV) and form insoluble hydroxides. This oxidation reaction is generally driven by the hydrogen peroxide generated from the cathodic ORR Eq. (5 )) [ 130 , 150 , 151 ] and can be illustrated by the Eqs. (6 )-( (8) [ 130 , 151 ]: This mode of operating of cerium salts resembles that of chromates, which could make them possible alternatives for chromatebased conversion coatings. However, the efficiency of cerium inhibition at lower pH has been questioned more than once [ 152 , 153 ]. Moreover, the fact remains that the price of REMS is high. The reason behind the high cost of REMS is not much related to their abundance on earth but more to the production process and their availability in the market [ 154 , 155 ]. Despite being relatively abundant on earth's crust (e.g. Ce is present at similar amounts as Cu but twice less than Cr), the REMS market is mainly controlled by the Chinese state-owned industries with production levels close to 85% (in 2015) of the world's supply and the imposition of strong restrictions (e.g. export quotas, taxes, prices, etc.), especially between the years 2008 and 2014 [ 154 , 156 ].
One other factor that may make the access to REMS difficult for corrosion protection purposed, is the high demands of other competing technologies that would also require considerable amounts of these elements [154] . All these listed factors make the industrial application of cerium or REMS for the sake of corrosion protection highly uncertain.

Zr/Ti based coatings
Zirconium and/or titanium-based conversion (Zr/TiCC) coatings have been receiving an increasing interest since their introduction as potential replacement for chromates and phosphates in the late 20th century [157] . Ever since, the technology has significantly progressed and is proposed in various commercially available product trademarks [ 157 , 158 ]. Although the main application of these Zr/TiCCs are used for the automotive bodies composed of the 5xxx [ 159 , 160 ] and 6xxx [161][162][163][164] series of Al alloys [165] , they are also highly suited for other Al alloys such as AA2024 used in the aerospace industry [ 88 , 102 , 166 , 167 ].
Many studies were devoted to the understanding of the formation mechanism of Zr/TiCCs and the influence of the Al alloy composition on it [ 161 , 163 , 164 , 166 , 168 ]. In a recent review by Milošev et al. [157] , the authors divided the formation of Zr/TiCC on Al alloys into different stages. The first stage is the dissolution of the Al native oxide layer activated by the presence of aggressive hexafluorometallate complexes (e.g. H 2 ZrF 6 , Na 2 ZrF 6 , H 2 TiF 6 , etc.), in the bath at low pH (2.8 to 4). This is then followed by the formation of hexafluoroaluminate compounds according to Eqs. (9) and (10) . The second stage involves the precipitation of the of Zr/Ti metal oxide layers driven by the increase of the local pH (to approx. 8.5) initiated at the cathodic sites (see Eqs. (11 ) to (14) ) [157] .
Stage I: Stage II: Besides their good corrosion resistance and remarkable adhesion performance to the metallic substrate, the use of Zr/TiCCs is energy and cost-effective as the formation process requires less time and lower temperatures in comparison to other conversion coatings [157] . Moreover, less sludge is induced from the preparation bath (low concentration of ionic species) and the involved compounds during the whole operation are classified as noncarcinogenic and non-eutrophic [169] .
Although the list of advantages in regard to Zr/TiCCs is quite relevant, there is still a large room for improvement, starting from the fact that these coatings are not self-healing but only barrier protective [157] and second, the Zr/Ti layer tends to be inhomogeneous [157] . One way to enhance the performance of the Zr/TiCCs is by the addition of inhibitors in form of inorganic and/or organic pigments. For the case of self-healing implementation, several studies have been published on the possibility of adding corrosion inhibitive agents such as vanadate [170] , cerium [171] and trivalent chromate [172] into the Zr/Ti formation bath for the corrosion protection of Al alloys. Similarly, a combination of additives such as poly-(4-vinylphenol and Mn 3 (PO 4 ) 2 have shown to favorize a homogeneous formation of the Zr/Ti-CC, when added into the reaction bath [ 173 , 174 ]. Both methods presented significant potential in improving the Zr/Ti-CC.

Alternative pigments in coatings
The list of potential substitutes to chromates does not end with the few examples mentioned in the previous sections, since when referring to an efficient corrosion protective system for Al alloys or any other metal alloy for that matter, there is no constraint on the chemistries or the nature of the system to be exploited. The only requirement is for the technology to be so-called green and REACH conform.
One can think of a multi-level protective system that can comprise both barrier and active corrosion protection functionalities.
Organic coating systems confer corrosion protection by acting as a barrier and preventing the mass transport of aggressive elements such as water and chloride into the metal surface [195] . However, this protection can easily be compromised upon mechanical damage or defects. Moreover, some organic coatings are very often prone to degradation due to their permeability to air and moisture. These issues were previously countered by the presence of active chromate pigments (e.g. SrCrO 4 ) in the coating formulation. Since their ban, few other technologies have been highlighted to optimize the anti-corrosion properties of the organic coating, when applied to Al alloy. The most common pathway is the direct addition of corrosion inhibitors into the coating formulation.
A few examples of direct addition of corrosion inhibitors in coating formulation used in Al alloy are given in Table 2 .
In addition to corrosion inhibitors, sacrificial metal pigments such as Mg, can also be added into primers to impart cathodic protection to the underlying Al alloy [207][208][209][210] . Mg having a more negative potential than the Al substrate will lead to the cathodic Table 2 Example of corrosion inhibitors incorporated directly into organic coatings for the corrosion protection of Al alloys.

Corrosion inhibitor
Al substrate Coating Ref.

Cerium
AA2024-T3 sol-gel and Epoxy [196][197][198][199][200]  sol-gel [ 198 , 202-205 ] 8-hydroxyquinoline AA2024 sol-gel [198] Tolyltriazole AA2024-T3 sol-gel [197] Chloranil (tetrachloro-p -benzoquinone) AA2017 sol-gel [206]  Mixed oxides/hydroxides Anionic [ 34 , 35 , 145 , 236 , 239 ] polarization of the latter and the dissolution of the Mg particulates. A study also showed the formation of a porous barrier Mg oxide layer on the Al substrate [207] . It is important to stress out that the addition of corrosion inhibitors and sacrificial metal pigments are just two methods among various others. The content of this section was restricted to these particular two types of pigments since they are more relevant to the current review.
Despite the promising results achieved with the addition of corrosion inhibitors or/and sacrificial particles into the organic coatings, these methodologies present a few disadvantages and need to be re-evaluated. For instance, in the corrosive media, a corrosion inhibitor to be incorporated needs to be have the right degree of solubility. A low solubility leads to insufficient self-healing properties whereas a high solubility results in a fast leaching of the inhibitor compromising the long-term active corrosion protection, along with a degradation of the coating due to possible delamination and blistering (caused by an increase of the osmotic pressure within the coating) [89] . The last effect was for example observed on Mg-rich primers applied on AA2024-T3 [ 211 , 212 ]. Moreover, the chemical nature of the corrosion inhibitor may also lead to its interaction with the polymer coating which will induce a loss of the inhibition property and degradation of the coating. The inhibition property of the corrosion inhibitor can also be lost due to its sensitivity to environmental factors. As a matter of fact, organic corrosion inhibitors such as 2-mercaptobenzothiazole and 1,2,3benzotriazole that were reported to inhibit corrosion activities on Al alloys, are subject to photodegradation [213] .This means that the direct incorporation of these inhibitors in a polymer coating may not provide the active corrosion performance that is needed.
Nevertheless, significant progress has been made to avoid some of the above-mentioned drawbacks. One key solution is to encapsulate the active agent using LDH nanocontainers, as it will be explained in the next section.

A nanocontainer based strategy
To avoid the direct incorporation of corrosion inhibitors into coatings and the limitations that can come by doing so, specific micro/nano delivery systems were designed and synthesized to host active agents such as corrosion inhibitors [ 77 , 89 , 90 , 214 , 215 ]. A wide variety of encapsulation approaches has been proposed in the past few years and were used for isolating/protecting the incorporating agents from the polymer resin, store the corrosion inhibitors and release them on demand when triggered by changes on the metal interface or upon mechanical damage. Some of these nanocontainer-based strategies were originally derived from other fields such as the food and pharmaceutical industry. A brief list of the most reported nanocontainer based strategies for corrosion protection, are given in Table 3 .
Three types of nanocontainers can be distinguished, according to the charge of the guest ions; the cationic exchangers, the anionic exchangers and neutral nanocontainers. Several triggering mechanisms can lead to the activation of these nanocontainers and the release of the intercalated inhibitor, namely: pH perturbations, presence of specific ions, change of the redox potential, mechanical rupture etc.
For instance, the cation-exchangers (e.g. zeolites and bentonites) may release the inhibitors in the presence of other cations in the interface (metal cations in the case of corrosion processes) [215] . For anion-exchangers (e.g. LDH), the aim is more directed to the capture of corrosive agents such as chlorides or sulfates followed by the release of the inhibitors. Finally, neutral nanocontainers rely on a desorption process and can have the advantage of hosting both anionic and cationic inhibitors.
The reason behind the choice of selecting LDH for this review, to the detriment of other nanocontainers can be explained by the following points; i) LDH have the remarkable option to be used both in form of anticorrosion pigments incorporated into a coating system [77] , as well as a conversion coating on the metal [ 48 , 84 ]. This means that it can be easily adapted according to different requirements. On one hand, LDH can be used as a powder/slurry if the wish is to confer additional active protection properties to a barrier coating by the addition of a self-healing functionality [ 15 , 18 , 55 , 89 ]. On the other hand, LDH can be applied directly to the metal surface as a conversion coating if there is a need for the first defensive layer (close to the metal) to be an active corrosion protective system [ 48 , 79 , 240 ]. ii) LDHs are among the most investigated and with more application potential anionexchange nanocontainers for "smart" active corrosion protection [ 9 , 13 , 18 , 48 , 81 , 84 , 241-249 ].
In the course of this review, we will be going through these two manners of using LDH and examine the latest progress in this area for the protection of aluminum alloys.

Synthesis of LDH pigments
The availability of LDH in their natural form is limited, but their preparation can be easily achieved in both laboratory and industrial scales at a moderate cost. Depending on the application, several methods were reported for the formation of LDH powders relying mostly on soft chemistry ( chimie douce ) reactions [ 2 , 45 , 250 ]. Some of these methods are; synthesis by co-precipitation , sol-gel method using alkoxides and/or acetylacetonate as starting precursors [ 39 , 251 , 252 ], urea hydrolysis method [253][254][255][256] , hydrothermal method [ 257 , 258 ], reformation [259] and mechanical milling [260][261][262] . Since the co-precipitation method is the most straightforward and commonly applied "one-pot" method, it will be given more attention in this review.
Typically, the synthesis by co-precipitation can be achieved at either variable pH (titration co-precipitation) or constant pH conditions. The latter option is preferable to obtain pure, crystalline LDHs. During the reaction, the pH of the solution is kept constant by the simultaneous addition of an alkaline solution (e.g. NaOH or NH 3 •H 2 O) together with the precursor solution of mixed metal salts (metals that will be part of the LDH). Usually, an alkaline solution is chosen according to the corresponding metal salts and the desired anion to be intercalated between the LDH galleries. Additionally, since it is difficult to avoid the presence of CO 2 in air, it is further advised to work under nitrogen or argon flow to avoid the formation of LDH intercalated with carbonates, if these are not desirable. The following equations were proposed as an example for the formation of ZnAl LDH pigment by co-precipitation method [263] : The LDH product of the synthesis by co-precipitation Eqs. (15 ) to ( (17) ) relies upon a crucial control of the pH of the reaction medium, the concentration and nature of both alkaline and metal precursor solutions (besides the molar ratio of the metal cation itself), the temperature and aging time. However, despite the above precautions, it remains very challenging to avoid the segregation of unwanted products (eg. oxide/hydroxide mixtures) [264] during the precipitation reaction, and this is mostly due to supersaturation and local increase of the pH. Therefore, it is advised to find a compromise between the different preparation conditions (see Fig. 4

a).
For instance, as a solution to the local pH increase, urea can be used to favor a homogeneous precipitation at a controlled pH [ 255 , 256 , 265 ]. This method leads to the formation of LDH with a high degree of crystallinity and larger flakes but with a narrower particle size distribution [255] . However, depending on the temperature, the urea procedure relies on a continuous release of ammonia and carbonate, generally yielding to the formation of carbonate-intercalated LDH [256] .
The above-mentioned modification of the synthesis bath, can intervene both in the nucleation and growth steps of the LDH. The conventional synthesis methods may help further optimize the LDH growth, but in most cases, they require a long synthesis time, high temperatures or pressure to obtain a pure-phased, well crystallized LDH that has a homogeneous oriented layer and a high specific area. Therefore, several assisted or/and post-synthesis treatment routes have been introduced for tuning the structural and textural properties of LDH powders, during the phase of growth (e.g. good dispersion and uniform distribution of the LDH particles), with the additional benefit of a reduced time and preconditioning ( Fig. 4 b). LDH synthesis using MWs (microwaves) allows to obtain an enhanced crystallinity together with considerable reduction of synthesis time [ 41 , 266-268 ]. This is achieved due to a well volumetrically distributed heating, which helps reduce thermal gradients. Indeed, in a study by Benito et al. [266] , SEM (scanning electron microscopy) micrographs of Mg-Al LDH have shown well-defined and very uniformly sized hexagonal platelets ( Fig. 5 b) in comparison with the Mg-Al LDH synthesized under constant pH conditions without microwaves treatment ( Fig. 5 a).
In addition to MWs treatment, ultrasound irradiation treatment was also applied to achieve a rapid synthesis and improve the crystallinity of the LDH phases. Compared to the traditional stirring, ultrasonic treatment allows a better dispersion of particles in solution. In this way, the reaction time is shortened and a more uniform LDH size distribution is obtained ( Fig. 5 c and d) [ 31 , 269-272 ].

LDH pigment anionic exchange for corrosion protection
Various pathways have been disclosed in order to obtain LDH pigments intercalated with the desired guest anion between the interlayers. Among the methods for intercalation/immobilization of the species in LDH, the most common ones are: (1) the direct intercalation during the synthesis by co-precipitation, as long as the precursors and the reaction medium contain the guest (and not contaminated by the presence of other anions susceptible to be intercalated) [83] , (2) the post ion-exchange reaction where the already formed LDH powder is immersed into a solution medium containing the salts of the desired guest anion and an anionexchange reaction takes place [ 12 , 273 ], and (3) the calcinationreconstruction procedure based on a so-called "memory effect". During heat treatment of the parent LDH (450 °C ≥ T), the interlayer water molecules are evaporated followed by a dihydroxylation coupled to transition metal oxide crystallization and the decomposition of the intercalated anions [ 274 , 275 ]. The mixed oxides can be brought into contact with an aqueous solution containing the anion to be intercalated; afterwards the LDH containing the desired anion can be reconstructed [276] . However, a prerequisite of this method is that the anion initially intercalated between the layers must be volatile and should decompose completely without forming any mixed compounds with the matrix cations [276] .
In the area of corrosion protection, most authors use the second method of post-synthesis treatment by anion-exchange reaction to obtain LDH with the aimed intercalated species.
The anionic exchange in LDH depends on several factors such as the electrostatic and H-bonding interactions between the mixedmetal hydroxide layers and the exchanging anions [277] . Nevertheless, similarly to the synthesis process of LDH, there are different factors that must be taken into account to promote a successful anion-exchange reaction. Amongst these factors is the affinity of the anion. Miyata et al. [278] claimed that the equilibrium constant for the exchange reactions increases when the radius of the bare anion decreases. The smaller the radius and the higher the charge densities of the anions, the more favorable is the exchange. Based on the calculation of various exchange reaction's equilibrium constant, an order of affinity of LDH to monovalent anions was established as following [ 279 , 280 ]: And for divalent anions as: Accordingly, since NO 3 − anion has a relatively low affinity, it can be easily replaced by other anions. Therefore, nitrate-containing LDHs are commonly used as parent compounds for subsequent intercalations (e.g. with corrosion inhibitors).
Despite the above assertions, it remains possible for low-charge and large anions (including organic species) to be intercalated. Indeed, other parameters involved in the anion-exchange reactions can be manipulated to facilitate the insertion of different anions. As a matter of fact, it was shown that some solvents such as methanol [281] and glycerol [282] can enlarge the space between the LDH interlayers allowing bigger species to be hosted [ 281 , 282 ].
Hibino [283] also highlighted the importance of the parent LDH crystallinity and the ratio of the metal cations of the LDH hydroxide layers, on the anion exchange selectivity. Choosing MgAl LDH as a model system, the following trends were established for a well-crystallized MgAl LDH: the higher is the Mg/Al molar ratio, the higher is the selectivity towards NO 3 − whereas the selectivity towards SO 4 2and Fis lower. The selectivity toward the different anions is less apparent for low-crystallinity LDHs [283] .
It should be pointed out that from a kinetic point of view, the anion exchange reaction is still subject to controversy. Next to the early studies by Miyata et al. [ 278 , 279 ] and Costa et al. [280] , several investigations were carried out to understand the mechanism and kinetics involved during the LDH anion exchange reaction es- pecially for bigger species. Meyn et al. [284] investigated the kinetics of the anion exchange reaction of five LDHs; ZnCr, ZnAl, MgAl, CaAl and LiAl LDHs, with several long chain organic species (carboxylic acid anions, arenesulfonates, secondary alkanesulfonates, alkynebenzensulfonates and ether sulfates). The parent LDHs were synthesized by co-precipitation and intercalated with nitrate. The anion-exchange was carried out by the immersion of the parent LDH into a solution containing the organic species to be intercalated. On one hand, it was found that the LDH is highly reactive to most organic anions characterized by a fast kinetic and a high exchange rate (80-100 %). Moreover, the rate and degree of reaction is generally independent of the chain length of the anion to be exchanged. On the other hand, the investigation showed that due to the small equivalent area of the LDH, the alkyl chain compounds (fatty acid anions, dicarboxylates, alkyl sulfates) are pushed to point away from the interlamellar surfaces and to form monomolecular films of high regularity [284] . Secondary alkanesulfonates and other specific anionic surfactants (mixtures of isomers and of compounds with different chain lengths) form wellordered bimolecular films of constant thickness, between the LDH interlayers [284] .
The mechanism of anion exchange of LDH with inorganic and/or organic species is of particular interest in the field of corrosion protection since it is a determinant factor to whether this novel technology can compete with formal Cr(VI)-based coatings. Indeed, the aptitude of LDH to host actives species such as corrosion inhibitors and promote self-healing/active corrosion protection, is the main reason for it to be considered as a suitable alternative. Therefore, several studies have been devoted to understanding mechanisms of ion-exchange of corrosion inhibitors [ 49 , 50 , 283 , 285 ]. For instance, Serdechnova et al. [50] , has investigated the anion exchange reaction of ZnAl-NO 3 and MgAl-NO 3 LDHs with two known corrosion inhibitors 2-mercaptobenzothiazole (MBT) and 1,2,3-benzotriazole (BTA). Using X-ray diffraction (XRD) as the main investigation tool, it was demonstrated that MBT can be easily exchanged and intercalated between the ZnAl LDH and MgAl LDH layers whereas BTA can only be intercalated within the MgAl LDH galleries. The anion-exchange reaction of NO 3 − with MBT/BTA was determined to be fast, without formation of intermediate phases and more interestingly with a rearrangement step where both organic species adopt a herringbone-like arrangement [50] .
Altogether, the intercalation of smaller and bigger corrosion inhibitors into LDH has proven to be very attainable. Recent works involving the corrosion protection of aluminum alloys reported the LDH intercalation with different inorganic inhibitors such as nitrate [ 48 , 264 ], vanadate [ 286 , 287 ] and phosphate [16] as well as organic inhibitors such as MBT [ 50 , 288 ], quinaldic acid (QA) and BTA [ 34 , 236 ], which were effective for AA2024 and other Al alloys (An overview list of the intercalated inhibitors into LDH will be provided in the next section). For the cited organic inhibitors, a prior deprotonation in NaOH was performed to convert the molecules into salts, so that they can be intercalated in the anionic form within the LDH galleries [50] .

Application of LDH pigments for corrosion protection
The use of LDH as functional additives in an effort to implement self-healing/active attributes in a multi-level corrosion protection system was addressed in several works [ 10 , 12 , 13 , 15 , 34 , 83 , 85 , 289 , 290 ]. In this particular case, a multilevel protective system refers to the combination of one or several corrosion preventive and/or protective mechanisms into one system [290] ( Fig. 6 ). It is possible to integrate these different functions into one coating systems (e.g. polymer) or distribute them within a frame composed of several layers of coatings. The last approach is generally what is used in industries such as the automotive [ 165 , 291 ]. Relevant features to the mechanism of active corrosion protection as well as the durability of the hosting coating matrix were highlighted, some of which are: -Efficient encapsulation of corrosion inhibitors, avoiding their direct contact with the coating's matrix, hence preventing both an unwanted deactivation and/or depletion of the inhibitor and the degradation of the coating [ 15 , 292 ]. -Triggered response to an external (e.g. scratch, mechanical damage) or internal stimuli (e.g. local change of pH) and release of the corrosion inhibitors. This also implies a controlled release over an extended period of time and/or a progressive release of the active agent into the affected area.
Buchheit et al . were the first to put forward a concept based on LDH as anticorrosion pigments in protective coatings. In a pioneering work published in 2003, Buchheit et al. [83] prepared ZnAl LDHs (referred to hydrotalcites) by co-precipitation of ZnCl 2 and AlCl 3 .6H 2 O precursors in a solution containing vanadate as the corrosion inhibitor in its decavanadate form (V 10 O 28 6 − ). The XRD patterns obtained for LDHs intercalated with decavanadates and chlorides are shown in Fig. 7 a. Accordingly, the layered structure of LDH is characterized by the reflections appearing at lower 2 θ values (between 10 °to 35 °) whereas the larger 2 θ values are associated to the structure within the mixed metal hydroxide layer in the compound ( Fig. 7 b). Therefore, during anion exchange reaction, a displacement of typical ( 003l ) reflections ( l = 1, 2, 3) can be observed. The anion exchange reaction of vanadate with chlorides within the LDH intergalleries will lead to the contraction of the latter and the formation of smaller crystallites of LDH-Cl. This phenomenon can be identified by the shift of the LDH containing vanadate reflections (003, 006 and 009) into lower angles ( Fig. 7 ).
The corrosion protection provided by vanadate-containing LDHs to aluminum alloy 2024-T3 was evaluated by incorporating the LDH particles (3 wt. %) into a polyvinyl alcohol-based coating (PVA) [83] . Both salt spray tests (SST) as well as electrochemical impedance spectroscopy (EIS) revealed significant improvement of corrosion behavior for AA2024-T3 plates coated with LDH-loaded coatings. Mahajanam et al. [32] demonstrated a few years later that in addition to decavanadate, Zn 2 + leached from the LDH structure upon mechanical damage can also contribute to the protection of AA2024-T3. This conclusion was made after the analysis of cathodic polarization curves in the presence of Zn 2 + . It was shown that a significant decrease of the diffusion limiting current density is observed when the concentration of Zn 2 + increases. This could be explained by the formation of a protective Zn(OH) 2 film [ 32 , 34 ].
Williams and McMurray promoted further the use of LDH pigments as a mean to impart AA2024-T3 with a resistance against filiform corrosion (FFC). Commercial LDH with a chemical formula of Mg 6 [Al 2 (OH) 16 ]CO 3 •4H 2 O were subjected to calcinationrehydration procedure in order to insert the corrosion inhibitors [ 12 , 13 , 85 ]. As a first proof of concept, LDH intercalated with nitrate, carbonate and chromate were incorporated into a polyvinyl butyral (PVB) coating and applied to AA2024-T3 panels. After visual examination and kinetics investigation by scanning kelvin probe (SKP), an effective hindrance to the propagation of FFC was observed for the AA2024-T3 LDH-containing coated panels. The FFC inhibition by LDH pigments occurred due to an uptake of aggressive Cl − anions, hence favoring an increase of the electrolyte pH (since the formation of HCl is unlikely after the capturing of Cl − ), which allows the repassivation of the AA2024-T3 surface [ 13 , 85 ]. At the same time, the inhibition effect depends highly on the intercalated inhibitor and increases in the following order: CO 3 2 − < NO 3 − ˂ CrO 4 2 − . In another study, the authors renewed the above experiment with a selection of organic inhibitors namely benzotriazolate, ethyl xanthate and oxalate. The resulting LDH pigments were also found to inhibit FFC in AA20204-T3, with the inhibiting efficiency increasing in the order of ethyl xanthate << oxalate < benzotriazolate [13] .
Kendig et al. [14] optimized the approach suggested by Williams and McMurray by using 2,5-dimercapto-1,3,4thiadiazolate as an organic corrosion inhibitor. The polarization tests were performed directly on a rotating Cu disk and have shown that this particular 2,5-dimercapto-1,3,4-thiadiazolate loaded LDH acted on inhibiting the oxygen reduction reaction (ORR). Moreover, similarly to strontium chromate, the inhibition effect of this organic inhibitor is irreversible which could imply that its mechanism of action relies on the formation of a possible protective layer. These results suggest that 2,5-dimercapto-1,3,4thiadiazolate loaded LDH could effectively inhibit filiform corrosion on AA2024-T3.
In 2010, Zheludkevich et al. [15] reconsidered the use of LDH intercalated with speciated vanadates (VO x − ). Two methods for LDH synthesis were tested: direct co-precipitation in a bath already containing the vanadate species and anion-exchange after synthesis by co-precipitation with nitrate-based precursors. Although a successful intercalation of vanadate was indicated by the XRD for both synthesis methods, vanadate-intercalated LDHs obtained by anion-exchange were found to be more crystalline than the ones obtained by direct co-precipitation. The LDH vanadate powders obtained by direct synthesis generated a relevant amount of amorphous aluminum vanadate phase, which had an immediate impact on the release behavior. Indeed, the release profiles of vanadate from ZnAl LDHs obtained by direct synthesis showed a lower concentration of vanadate released into the solution in comparison to the ones obtained by anion exchange. Moreover, the release profile of the LDH vanadate obtained by anion-exchange revealed a fast-occurring release during the first hours followed by a chemical equilibrium. It was also noted that the saturation limit for the vanadate release diminishes when more LDH-vanadate is added into the solution. The authors argued in this respect, that the exchange reaction between vanadate and chloride is chemical in nature and is dominated by an anion-exchange equilibrium rather that a solubility equilibrium.
A multilayer painting scheme was prepared by incorporating the above LDH-VO x pigments (prepared by direct synthesis and by anion-exchange) into an aeronautical based painting scheme and applied to AA2024 specimens [15] . The EIS curves together with their fitting models displayed a higher oxide barrier resistance than that of a conventional chromate-loaded paints. The coating resistance was also higher under prolonged immersion conditions in the presence of LDHs. In addition, standard tests were performed, including FFC and Q-panel condensation testing (QCT). While FFC testing revealed that LDHs led to better protection against FFC than chromate-loaded paint, QCT testing showed a higher number of blisters in the presence of LDHs when compared to reference (without any inhibitor additive) and chromate-doping systems [15] . A similar study on the corrosion protection efficiency of ZnAl LDH intercalated with vanadate was reported by Vega et al. [286] . Alkyd based primers with different LDH pigment concentration (5 %, 10 % and 15 % ZnAl LDH intercalated with vanadate) were prepared and their anticorrosion performance were compared to that of a chromate-based coating containing 10 % zinc chromate. The best results were obtained with the alkyd primer containing 10 % ZnAl LDH with vanadate, and confirmed the high potential of ZnAl LDH intercalated with vanadate to replace chromate-based systems. However, much care must be given to the content and dispersion of the pigment in a polymer matrix so that they do not interfere with other coating properties such as adhesion.
Although efficient for the corrosion inhibition of AA2024, vanadate was not the only candidate proposed to be intercalated into LDH pigments and used to enhance the anticorrosive properties of polymer coatings. Other groups proposed the use of organic corrosion inhibitors. For instance, Poznyak et al. [34] investigated the use of MBT and QA as possible organic inhibitors to be loaded into ZnAl LDH and MgAl LDH nanocontainers for AA2024 corrosion protection. The different LDHs loaded with organic inhibitors were added to a solution of 0.05 M NaCl (50mg of LDH in 10ml solution) and their anticorrosion performance were tested on AA2024-T3 by EIS. The best corrosion performance was demonstrated with the LDH loaded with MBT. Optical micrographs of the AA2024-T3 samples exposed to the above solution for 2 weeks showed the formation of a protective layer with samples exposed to solutions with LDH loaded with QA and MBT. However, the layer formed from the LDH loaded with MBT presented a much denser and compact structure. This formed compact layer allows to provide a long-term corrosion protection to the AA2024-T3 panels.
Over the years, many studies examined the possibility of intercalating bigger organic corrosion inhibitors for the protection of aluminum alloys were published. For instance, Stimpfling et al. have explored through various studies a number of organic inhibitors to be intercalated within the LDH intergalleries. Among them 2-hydroxyethyl phosphate (2-HEP) [54] , ethylenediaminetetraacetic acid (EDTA) [55] , organo-modified compounds by aniline and benzene derivatives [17] as well as various α-amino acid molecules [56] . In all mentioned studies, the different produced LDHs loaded with the organic inhibitors were incorporated into a primer polymer coating and were tested by DC polarization and EIS techniques. They have all shown viability as potential self-healing corrosion protective systems for AA2024.
An attractive feature of encapsulation of corrosion inhibitors in nanocontainers is the possibility to combine different inhibiting species in the same coating system. Tedim et al. [16] examined the possibility of combining LDHs loaded with different inhibitors in the intention of achieving synergistic effect for active corrosion protection of AA2024. The inorganic inhibitors vanadate (VO x n − ) and phosphate (H x PO 4 n − ) together with the organic inhibitor MBT, were chosen to be loaded separately into LDH and the mix was incorporated in a polymer matrix (ratio 1:1 and total amount of dry LDH was 10 %) or directly dispersed in the electrolyte (ration 1:1 and total concentration of LDH was 5 g.L −1 ). In the latter case, EIS results showed that the combination of vanadate-containing LDH with MBT-or phosphate-containing LDHs revealed a clear synergistic effect. Once the LDH pigments were added into the coating, the synergistic effect was limited due to the interaction between the LDHs and the polymer matrix. The best corrosion performance was observed for the system constituted by LDH-MBT added in solgel pre-treatment + LDH-VO x in the primer. Both coating and oxide resistances obtained from EIS fitting showed the highest values among the investigated systems. The proximity of the LDH-MBT to the substrate provides a short protection while LDH-VO x present in the primer can provide a long-term protection.
In a similar experiment, Subasri et al . [18] prepared Zn-Al LDH pigments loaded separately with vanadate, MBT, molybdate, phytic acid and 8-hydroxyquinoline (8-HQ). The loaded LDH were then taken and dispersed individually in a hybrid sol-gel silica matrix. The obtained LDH-containing sol-gel silica coatings were applied to AA2024-T3 substrates according to a bilayer configuration, with the first sol-gel layer containing LDH loaded with either MBT, molybdate, phytic acid or 8-HQ and the second layer being in every case silica with LDH-vanadate. EIS, DC polarization and SST assessment of the different configurations showed enhanced corrosion resistance even when compared to chromate conversion coated samples owing to a barrier protection combined with the triggered response from the release of the corrosion inhibitors. The best corrosion performance was achieved with the bilayer configuration containing LDH-vanadate/LDH-molybdate. Besides, no interference with the adhesion strength was found for all five systems.
The idea of mixing inhibitors has been also applied to prevent the galvanic corrosion of aluminum alloy (AA6061) coupled with carbon fiber reinforced plastic (CFRP). Serdechnova et al. [236] exploited a combination of two nanocontainers, one being MgAl LDH loaded with BTA and the other was bentonite impregnated with cerium. Fig. 8 a and b represent the specimen design, which illustrates the AA6061 alloy galvanically-coupled with CFRP and embedded in an inert epoxy resin. The sample was coated with a commercially available bi-component epoxy resin containing the two nanocontainers. The self-healing ability of this model specimen was tested by applying two artificially made needle-like defects, located in both materials of the AA6061 + CFRP galvanic couple. The activities on the surface with immersion time was mon-itored by scanning vibrating electrode technique (SVET) ( Fig. 8 c  and d). Anodic and cathodic current densities are well defined and located strictly at the artificial defect zones over AA6061 and CFRP in both cases at the beginning of immersion [293] . The blank system demonstrated relatively stable values of corrosion currents in the defects during the measurement time of 24 hours, while a well-defined self-healing effect is observed in the case of the coating impregnated with nanocontainers.
The above cited works prompt one to think on possibilities of using rare earth compounds such as cerium for the corrosion protection of Al alloys. In this regards many attempts were done to introduce rare earth compounds into the LDH structure, most of which involve the rare earth cations (e.g. Ce 3 + , La 3 + etc.) to be adsorbed or directly integrated into the brucite-like sheet of the LDH layers and this mostly for other application than corrosion [ 40 , 273 , 294-297 ].
Nevertheless, the procedure used to benefit from rare earth and LDH together in one framework is not straightforward owing to the fact that they are positively charged (not possible to intercalate) and have a large ionic radius (difficulties to insert in the mixedmetal composition of the double hydroxide layers). However, some strategies mostly focusing on cerium, were developed for the corrosion protection of AA2024.
Liu et al. [298] prepared ZnAlCe-LDH powders with different Ce 3 + / (Al 3 + + Ce 3 + ) ratios by the co-precipitation method. The XRD results showed an increase of the d-spacing to higher values proportionally to the content of cerium. The authors argued that this increase is proof of successful insertion of Ce 3 + within the hydroxide sheets. It was demonstrated that the best corrosion performance on AA2024 was attained using the ZnAlCe-LDH with the above ratio of 0.1 because of a synergistic mixture of ZnAlCe-LDHs and CeO 2 nanoparticles. Relying on a different method, Carneiro et al. [51] managed to build an optimized LDH framework through a layer by layer (LbL) approach in order to allow the hosting of two type of corrosion inhibitors (MBT + Ce 3 + ). The MBT organic anion was loaded into the LDH galleries whereas the Ce 3 + was hosted between the PSS/PAH layers (see Fig. 9 ). This concept allows to obtain a synergistic inhibition effect by the combination of MBT and Ce 3 + within one nanocontainer.
Electrochemical studies based on EIS and direct current (DC) polarization on AA2024 coated with a sol-gel coating containing these LbL modified LDHs (0.5 wt. %), revealed an improvement of the corrosion resistance properties in comparison to the reference coating and the one loaded with just LDH-MBT. This behavior was associated to the poor interaction of MBT (which may be released at an early stage of coating preparation) with the sol-gel matrix, when LDHs are not covered with polyelectrolytes. Therefore, the additional polyelectrolyte treatment allowed not only to integrate cerium inhibitors into the LDH framework but also to promote compatibility between the pigments and the polymer matrix.
Aside from offering the possibility of a "smart" active corrosion protection, LDH pigments can also play a role in the prevention of corrosion. In some of the above reviewed studies, the nano-trapping and sensing aspects of LDH nanocontainers has often been mentioned [ 34 , 83 , 238 ]. These two features are often related since the first action, nano-trapping, is an indication of changes happening at the interface of the Al alloy which is translated as a sensing property. This was emphasized in a work by Buchheit et al. [83] . According to this investigation, a change in the basal spacing of LDH would mean a nucleation of a new LDH phase containing the corrosive species chlorides or sulfate. This change in basal spacing can be uncovered by XRD analysis and would be a primary indication of electrolyte permeation in the coating.
The entrapment of aggressive ions such as chlorides was also addressed in more details in the study by Tedim et al. [35] . ZnAl LDH loaded with nitrate and chlorides were prepared and incorporated into an aliphatic-and acrylic-based polyurethane (automotive coating) and were subjected to permeability tests in a solution of 0.5 M NaCl. It was found that the permeability of the coatings is 20 times lower in the presence of ZnAl LDH loaded with nitrates than it was in the presence of ZnAl loaded with chloride or without LDH. This positive effect was associated to the entrapment of the chloride by the ZnAl LDH nitrate through an anion exchange process. This was demonstrated by the shift of the XRD reflection of the ZnAl LDH nitrate to higher angles, due to the substitution of nitrate with chlorides between the LDH layers. After a month of measurement, 70 % of the nitrate were exchanged with chloride ions.
In a different manner, the sensing functionality of LDH was addressed by Wong et al. [289] . The structural memory effect of Li 2 [Al 2 (OH) 6 ] 2 CO 3 •nH 2 O LDH powder was utilized to detect the water uptake in optically opaque organic coatings in a remote and non-destructive manner. The LDH was synthesized by coprecipitation and calcined in air (220 °C < T) and the resulting LiAl mixed hydrated oxide powder were incorporated into a commercial epoxy resin. The coating was applied on AA2024-T3 plates and exposed to 0.5 M NaCl solution for different periods. Using XRD as the main investigating tool, the rehydration transformation of the calcined LDH could be monitored and quantified by evaluating the peak height ratio (PHR) of the LDH characteristic reflection (003). Water-uptake is generally considered as the first step of more serious consequences for metal corrosion such as blisters, delamination and adhesion loss. This approach is suggested to prevent this from happening by an efficient and early detection of water-uptake in organic coatings.
As highlighted in the earlier sections, LDH nanocontainers have been exploited in different ways to offer extrinsic self-healing Fig. 10. Scheme summarizing a few challenges that may be encountered when using LDH nanocontainers as additives in a polymer matrix (modified from [293] ).
functionalities (see Table 3 ). They can be integrated in frame of a much complex multi-level corrosion protective system. However, one would think that using LDH as pigments would still lead to the encountering of some of the issues discussed in Section 2.3 and described in Fig. 10 . However, through the previous sections, we learned that LDH can be tuned to fit the various requirements in order to obtain a reliable active corrosion protective system.
For instance, osmotic blistering that is caused because of a concentration difference across the coating membrane, can be prevented by controlling the amount of inhibitor loaded into the nanocontainer as well as the amount of LDH nanocontainers hosting the so-called inhibitor. This could also contribute for the reduction of coating disbonding and delamination. There are established testing and monitoring methodologies that can be followed to obtain an optimized coating system. For instance, Mahajanam and Buchheit managed to define a critical percentage/concentration of LDH pigments below which blistering starts to take place for a primer [299] . This was achieved by visual examination of epoxy primers containing different LDH concentration, applied into glass substrates and immersed them into an aggressive media. Zheludkevich et al. relied on Q-Panel condensation test (QCT) (ASTM D 4585-99) to evaluate the degree of blistering caused by the incorporation of LDH pigments into a water-based epoxy primer [15] . Chico et al. compared the degree of blistering of LDH, chromate and Ca/Si pigments added into an alkyd coating based on ASTM D714-56. Vega et al. [286] tested and evaluated different LDH/coating ratios using the same standard.
The issue of compatibility between the LDH pigment and the polymer can be overcome by modification of their outer surface using surfactants or other organic compounds. Studies have reported the modification of the LDH surface to achieve different properties. Just to cite a few examples; LDH was modified with cetyl trimethyl ammonium bromide (CTAB) and octadecyl trimethyl ammonium bromide (ODTMA) surfactants to obtain better photocatalytic properties [300] . In another work, the hydroxyl groups on a MgAl LDH surface were altered by the introduction of amine groups to help improve the LDH drug delivery property by enhancing their compatibility with blood [ 301 , 302 ] but also their physical adsorption of CO 2 , when it comes to environmental application [303] . MgAl LDH surface could also be changed from hydrophilic to hydrophobic to enhance flammability properties using different surfactants [304] . For anti-corrosion applications, Hu et al. [305] reported the possibility of chemically grafting ZnAl LDH nanocontainers loaded with vanadate in order to make it compatible for incorporation in a commercial epoxy coating. Briefly, the ZnAl LDH-nitrate was prepared by co-precipitation followed by an anion-exchange reaction to intercalate vanadate. The obtained ZnAl LDH loaded with decavanadate was then subjected to a hydroxylated treatment before its immersion in a solution of γaminopropyltriethoxysilane (APTS) to form a bonding layer at the surface of the LDH. The final step consisted of an in-situ polymerization of polyaniline (PANI) in the presence of the APTS modified ZnAl LDH to produce a polymer-clay composite PANI/ZnAl LDH loaded with decavanadate (see Fig. 11 ). The resulting surface engineered LDH could easily be incorporated in an epoxy based coating and offer efficient corrosion protection to mild steel substrate [305] . A similar surface modification was also attained with the work of Du et al. [306] , where camphorsulfonic acid doped polyaniline(CPANI)-modified MgAl LDH composites were fabricated through an oxidative polymerization process. The altered LDH pigments allowed not only to ensure compatibility between the LDH pigment and the waterborne epoxy resin, but also to allow a better dispersion of the LDH and avoid the probability of agglomeration. Although this coating presented no active corrosion protective functionalities (no inhibitor loaded into LDH), it did exhibit superior barrier protection to the mild steel.
Industrial coating systems such as primers are generally composed of several different components and fillers that enhance the properties of the coating and guarantees a long-lasting use. For instance, different types of surfactant can be found on a paint system such as: i) dispersing agents that promote the dispersion of solid particles and prevent their flocculation/segregation through charge stabilization and, ii) wetting agents that help reduce the interfacial tension at the solid-liquid interface (e.g. between a solid pigment and the liquid polymer coating) enables the water to wet the pigment particles, hence allowing the good insertion of the lat- ter. Some wetting surfactants can also facilitate the application of the coating on a substrate, thus improving adhesion [307] . Aside from the additives in the coating itself, the mixing process of the pigments and polymer matrix is generally achieved with sophisticated industrial mixing tools or agitators that contribute greatly in the optimization of the final coating formulation. Consequently, all these parameters should be taken into consideration as they can significantly facilitate the incorporation of LDH pigments into an organic coating.
It should be emphasized that, to this day, the reported studies on LDH nanocontainers and their incorporation in an organic coating, do not show an important diversity in terms of the selected coatings. It is important to perform more systematic and comparative investigations to find a range of formulations that are most suitable for the different available LDH pigments. Moreover, some of these investigations should include mechanistic and kinetic understanding of the inhibitor release from the LDH nanocontainers. The selected coating formulations should not interfere with the inhibitor release characteristics of the LDH nanocontainers but instead, it should allow an efficient diffusion of the released inhibitors into the unprotected area. It has been recently shown that the leaching kinetics of corrosion inhibitors and their diffusion and transport through the coating matrix are two independent phenomena. Accordingly, the leaching of the corrosion inhibitors starts from a cluster of connected inhibitors pigments and proceeds until no more species are left in the clusters. The removal of the connected inhibitor leads to the formation of a percolating network of connected voids that will enable the easy transport of further corrosion inhibitor species [308] . Therefore, when it comes to LDH pigments, one can suppose that the first stage related to the release of inhibitor from the LDH pigments, depends on the trigger and the solubility of the inhibitor whereas, the second stage that is the leaching process over time, is controlled by the creation of a percolating network of voids that will allow further transport of the inhibitors [308][309][310][311] . The described model was summarized into a parameter called the "percolation threshold", which is the critical value of inhibitor pigment volume concentration (PVC) that needs to be considered during a coating formulation [310] . However, this model above was established on the ba-sis of a direct addition of inhibitors into the coating formulation and it is expected that the mechanism may differ for an encapsulated inhibitor. Therefore, for the coating formulation containing LDH loaded with corrosion inhibitors, the percolation threshold should be defined by considering the following factors: the percentage of inhibitor loaded into the LDH, the volume of the final LDH loaded with inhibitors, the efficiency and dissolution of the corrosion inhibitor, the release kinetics of the inhibitors from the LDH.
Adjusting the model to a system based on LDH encapsulated inhibitor is key for the smooth integration of these nanocontainers into industry.
Through the previous sections, it could be undoubtedly confirmed that the use of LDH nanocontainers as "smart" carriers for active agents in a coating framework, can be considered as a potent strategy to replace Cr(VI) pigment-based systems. The LDH pigments can be prepared through various methods with a chemical composition that can be easily varied according to the needs. They can host several corrosion inhibitors, organic and inorganic inhibitors either in frame of the same LDH nanocontainer or as a combination of several LDHs containing different inhibitors. It can be surface engineered to allow the intercalation of a diversity of corrosion inhibitors (e.g anions and cations inhibitors in an LbL framework), or surface modified to promote compatibility within a coating formulation. It offers protection to the hosted corrosion inhibitor and allows its controlled release when triggered by specific factors (e.g. pH, chlorides…). These are just a few advantages among others. A summery of the reviewed studies in this section are represented in Table 4 .
LDH pigments may be regarded as possible alternatives to Cr(VI) pigments but not to chromate-based conversion coatings (CCC). This fact is simply due to the distinct role played by an organic coating containing active pigments and a conversion layer directly generated on the surface of an Al substrate. A conversion coating does not only improve the corrosion resistance, but also strengthens the adhesion to subsequent organic coatings. As a direct alternative to CCC, LDH conversion coating (LDH-CC) is suggested . And it offers as much advantages as presented for the LDH as pigments. This will be reviewed in the following sections.

LDH as a functional conversion layer
In previous sections, studies on the use of LDH as pigments to improve corrosion protection of different hybrid and organic coatings were reviewed and, in some cases, their performance would match or exceed chromate-based pigments. However, similarly to Cr(VI) technology, LDH can also be directly applied to the metal surface in form of a conversion layer [ 10 , 33 , 38 , 48 ]. Keeping in mind that most well-known and studied LDHs are based on compositions such as ZnAl or MgAl, attempts to develop LDHbased conversion films have occurred during the last 25 years [ 8 , 10 , 11 , 48 , 86 , 240 , 244 , 312-317 ]. On one hand, this work can be considered as a new field with methodologies competing with chrome-free conversion treatments. On the other hand, it is an extension and improvement of already existing processes such as the sealing steps in anodizing processes.
Just like the conversion coatings described in Section 2.2 , LDH conversion layers can be prepared directly on the surface of the Al substrate through a chemical/electrochemical conversion process. The active corrosion protective property is implemented by the intercalation of corrosion inhibitors between the LDH-CC galleries. In the event of mechanical damage or changes the interface (e.g. pH, presence of chlorides or/and sulfate, etc.), the LDH-CC is triggered and a subsequent release of the corrosion inhibitors takes place, followed by the entrapment of aggressive species. Hence, impeding the corrosion activity at the affected zone ( Fig. 12 ).
A comprehensive understanding of the LDH-CC synthesis and mechanism of corrosion protection of Al alloys, is provided in the next sections.

LDH-CC Synthesis
From the time of the first reported achievement of LDH as conversion coatings for Al alloys [8][9][10][11] , studies on the optimization and modification of the morphology and properties of these LDH films have been constantly increasing. According to the recent review by Guo et al [63] , two main pathways can be adopted to obtain LDH-films, by physical deposition or in-situ formation.
Physical deposition methods include Layer-by-layer (LbL) [ 241 , 318-321 ], solvent/colloid evaporation [322][323][324][325] , and sol-gel spin-coating techniques [ 63 , 326 ] . Although these physical deposition methods allowed the preparation of diverse LDH films with interesting properties, they have not been relevant for anti-corrosion applications. This may be associated with the fact that physical deposition does not allow an intimate connection between the substrate and the LDH film. Indeed, the substrate does not serve as precursor. This means that the adhesion is weak which clearly is an obstacle for corrosion protection purposes. Moreover, the shape of the substrate could be an issue since with deposition methods, it may be challenging to cover surfaces with complicated shapes [ 63 , 327 ].
Contrary to physical deposition methods, in-situ LDH preparation allows an immediate contact of the LDH film layer and the Al substrate. In-situ LDH growth can be achieved either by electrodeposition or by co-precipitation treatment.
In-situ growth by electrodeposition is fairly known [ 33 , 38 , 63 , 312 , 326 ]. Most reported studies state that the electrodeposition of LDH film is based on the reduction of nitrate ions and the generation of OH − ions (see Eq. (14 )), which will lead to an increase of the local pH of the working electrode namely the substrate to be treated [33] .
This increase of pH drives the precipitation and formation of the LDH film on the substrate. However, the substrate to be treated is not the source of the cations involved on the LDH film growth. This is why various conducting materials such as Pt [ 33 , 43 , 328 ], Au [33] , glassy carbon [329] and Ni [330] could be treated with LDH films. This approach is more adequate for the area of energy conversion and storage [ 70 , 71 ] and has not been adopted for corrosion protection purposes, up to now. The reason might be that the resulting LDH has a disordered orientation with low crystallinity and purity due to the formation of unwanted intermediate phases (eg. metal hydroxides/oxides). This could interrupt further growth and thickening of the LDH film during the preparation process. The other important drawback of this method is the poor adhesion of the formed LDH film to the metallic substrate [331] .
The second and most known pathway for in-situ LDH growth is by co-precipitation, which allows the formation of 2D perpendicular oriented LDH flakes [326] . From a chemical point of view, insitu LDH growth is considered more advantageous since it involves the formation of chemical bonds, which strengthen the adhesion of the film. Moreover, in-situ LDH film formation is a one-step process and occurs independently of the shape of the substrate. This means that this method can be applied to a large area of application [63] . The simplest in-situ LDH growth can be achieved by a co-precipitation process , which is an extension of the LDH powder synthesis by co-precipitation. In this case, the substrate to be treated is generally also one of the precursors. In the case of Al substrate, M/Al LDH films can be fabricated by the immersion of the substrate in a bath containing a metal cation M 2 + /M + (Zn 2 + , Mg 2 + , Li + ...etc.) precursor in certain conditions (pH, temperature, concentration etc.) while the Al 3 + ions are generated by the dis-solution of the Al substrate. Since this process is derived from the co-precipitation method for LDH powder synthesis, most facts associated with the former method apply for the LDH film formation. In other words, the synthesis can be achieved by using urea [312] , ammonia [314] or simply sodium hydroxide [44] as precipitating agents. However, as mentioned in the first section of this review, the final morphology of the 2D LDH flakes will vary according to the used precipitating agent. About corrosion application, the coprecipitation synthesis using ammonia or sodium hydroxide is generally the method of choice. The reason being the same as for LDH synthesis as pigments, the urea hydrolysis tends to release carbonates that will in turn be intercalated into the LDH galleries which is unwanted (since carbonate cannot be easily exchanged).
In the following sections, most of the LDH functional layers were prepared by in-situ hydrothermal/co-precipitation method.

LDH-CC anionic exchange for corrosion protection
Contrary to LDH nanocontainers in form of pigments, there has not been many studies in terms of intercalation possibilities for LDH in form of conversion coatings especially in the area of corrosion protection. In the context of corrosion protection, vanadate was the most utilized model corrosion inhibitor. First, because vanadate is a known efficient inhibitor for the corrosion protection of Al alloys [ 92 , 332 , 333 ] and second, the successful intercalation of vanadate anions between the LDH layers has been evidenced throughout several investigations [ 44 , 48 , 286 , 287 , 314 ]. Other corrosion inhibitors namely molybdate (MoO 4 2 − ) [52] , MBT [ 52 , 334 ] and 8HQ [335] have also been claimed as potential candidates to be intercalated between the LDH layers.
So far, the intercalation process of the cited species was carried out through anion-exchange reaction by simple immersion of the LDH treated Al alloy into a solution containing the respective species at a specific pH, concentration and temperature. Most of these conditions were derived from the knowledge acquired from the anion-exchange reactions carried out with LDH in form of pigments [ 18 , 19 , 50 , 288 ]. This also includes the preference of selecting LDH intercalated with nitrate as a parent LDH, since similarly to the LDH pigments, LDH films intercalated with OH − and CO 3 2 − are challenging to exchange [ 79 , 334 ].
One can suppose that the lack of possibilities in terms of species that could be intercalated in the LDH-CC can be associated to the inelastic nature of the LDH film lamella. Indeed, as shown in earlier sections, a wide range of inorganic and organic species could be intercalated into LDH powders and most argued that it was due to the nature of the LDH pigments to expand and host even bigger molecules. In the case of LDH-CC a strain may exist, since the LDH films are chemically bonded into the treated metal substrate. Hence, two possible scenarios can be predicted; (i) the bigger molecules may cause a forced expansion of the LDH flakes that will be accompanied by a fragmentation of the LDH crystallites [ 334 , 336 ] or, (ii) the LDH will not expand and the species do not intercalate between the layers. For instance, Neves et al. [334] noticed that after anion exchange reaction from NO 3 − to MBT − in ZnAl LDH-CC the XRD reflections (003) and (006) corresponding to the LDH-NO 3 phase have disappeared and a new peak has appeared at lower angles (to the left) ( Fig. 13 b). This is in agreement with the peak shift observed with ZnAl LDH in form of powders ( Fig. 13 a). Moreover the (001) and (113) reflections associated to the LDH hydroxide structure have not changed after anion-exchange reaction. This was enough to prove the successful intercalation of the organic corrosion inhibitor ( Fig. 13 b).
However, when comparing XRD patterns obtained with ZnAl LDH intercalated with MBT − in form of pigment ( Fig. 13 a) and in form of conversion coatings ( Fig. 13 b), it can be disputed that the amount of LDH phase, in the case of LDH-CC, after the anion exchange process with MBT − is much lower and less crystalline in comparison to the parent LDH-CC with NO 3 − . The latter finding was explained by the fact that since LDH-CC are attached to the metal surface, there is a high probability for the LDH film to be fragmentated [334] .
The hypothesis of possible fragmentation was also suggested in a recent work by Bouali et al. [248] . The authors monitored the anion exchange reaction (NO 3 − → Cl − ) of ZnAl LDH film grown on AA2024 and zinc substrate to track the kinetics differences. The measurements were carried out in in-situ mode, using synchrotron high performance X-ray diffraction. In both cases, when ZnAl LDH is grown on AA2024 and pure zinc, an important amorphous phase was noticed on the XRD patterns which was associated to a decomposition of the LDH crystallite as a result of the exchange reaction. This is an interesting fact, considering that Cl − anions are much smaller than both NO 3 − and MBT − . In other words, an expansion or contraction of the LDH galleries may always induce stress leading to a partial or a total fragmentation of the LDH crystallites. The study also revealed some other characteristics of the anion placement between the LDH galleries. For instance, it is claimed that the nitrate anions are positioned in a 70 °angle, perpendicular to the cationic layers to achieve the highest compensation of the charge. After intercalation with chlorides, the basal spacing is reduced, followed by a decrease of the LDH crystallite size [248] . The latter effect is more pronounced for the LDH grown on pure zinc. Although this review focused on LDH films produced on Al alloys, it is important to point out that this comparative investigation showed that the kinetics of exchange reaction (NO 3 − → Cl − ) is faster for the LDH films grown on pure zinc than the ones on AA2024. Moreover, the resulting interlayer arrangement of the LDH-Cl is dissimilar in both cases. This entails that the starting substrate for LDH growth may greatly influence the kinetics of anion-exchange reactions.
Based on the same investigation tools of the above cited work [248] , Iuzviuk et al. [317] managed to establish a kinetic trend of anion-exchange for a few corrosion relevant species namely Cl − , SO 4 2 − and VO x n − . The parent ZnAl LDH film intercalated with nitrates were produced on a pure zinc substrate. Three exchange reactions were monitored: ZnAl LDH-NO 3 → ZnAl LDH-SO 4 , ZnAl LDH-NO 3 → ZnAl LDH-Cl and ZnAl LDH-NO 3 → ZnA LDH-VO x . It was found that the first two reactions occur faster than the latter reaction. Based on an Avrami-Erofeev (AE) analysis, it was found that the above anion-exchange processes happen in two stages, the first is a two-dimensional diffusion-controlled reaction, and the second is characterized by a one-dimensional diffusion-controlled reaction. Both stages include a decelerator nucleation effect. The last anion-exchange reaction Zn-LDH-NO 3 → Zn-LDH-VO x is governed by a slow reaction , which the authors attributed to an influence of the vanadate speciation characteristics. Indeed, two LDH structures were identified that were attributed to the ZnAl LDHs intercalated with V 4 O 12 4and V 2 O 7 4-. These polyvanadates were claimed to be promoting efficient corrosion inhibition of Al alloys [44] . The results of their kinetic study helped define a sequence of priority for anion-exchange reactions, with the parent ZnAl LDH -NO 3: As mentioned before, the starting substrate may affect the kinetics of anion exchange reaction [248] therefore it might be in- correct to extend the finding from [317] into LDH film formed on Al alloys.
Most known kinetics and modelling studies have been carried out on LDH in form of powder. For example, the work by Miyata et al. [ 278 , 279 ] and Costa et al. [280] are considered as basics in terms of LDH anion affinity and exchange reaction. However, till recently there has not been a clear study that asserts that the knowledge can be extended to LDH conversion coatings.
When dealing with LDH films, there are more challenging aspects that need to be taken into account such as the nature of the substrate and the induced stress from the expansion/contraction of the LDH lamellae during the exchange reaction. The latter aspect concern mostly in-situ LDH grown film. For films produced by deposition pathways (LbL, solvent evaporation or spin coating), the restrictions may be different since the resulting LDH is not attached to the metal substrates.

LDH conversion coatings on aluminum alloys
The first reported studies on surface modification of aluminum alloys by LDH films for the sake of corrosion protection, appeared earlier than that of LDH pigments. Indeed, before the group of Buchheit and co-workers reported their studies on LDH as pigments for corrosion protection, they first promoted the use of LDH as functional surface treatment for the corrosion protection of Al alloy in 1994 [8] .
Buchheit et al. demonstrated a formation of what they called a talc film formation on four different Al alloys (AA1100, AA2024-T3, AA6061-T6, and AA7075-T6), when immersed in an alkaline Licontaining salt bath. Their investigation revealed that this talc film had the typical structure and stoichiometry of the hydrotalcite-like compound Li 2 [Al 2 (OH) 6 ] 2 •CO 3 •nH 2 O and provided an efficient corrosion resistance towards standard salt spray testing [8] . In 2002, three more studies by the same group were published with each revealing more about the likelihood of these nanocontainers to re-place the toxic CCC [9][10][11] . For instance, the authors claimed that the development of these systems was done with a reduced number of processing steps using non-toxic chemicals. Moreover, the requirements essential to produce these LiAl LDHs on Al substrates are the following: (i) dissolution of the native Al-oxide, (ii) sufficient concentrations of the reactants in order to cause precipitation, (iii) in a reasonable short time of coating formation. To accelerate the LDH film formation [11] , it was shown that the addition of oxidizing agents such as hydrogen peroxide into the hydrotalcite baths accelerated the dissolution of the Al-oxide and the formation of the LDH-CC. Even better, the resulting LDH coatings presented superior corrosion resistance.
Several years after the first studies of Buchheit et al., Guo et al. published in 2009, a novel method to produce ZnAl LDH films on a pure aluminum sheet (purity > 99.5 wt. %) [337] . The authors proposed a one-step co-precipitation methodology based on the immersion of the Al sheets in a bath containing zinc nitrates and ammonium nitrate salts adjusted to a pH of 6.5 by means of a 1 wt. % ammonia. The obtained compact ZnAl LDH layer (see Fig. 14 ), showed not only a good adhesion performance but also remarkable corrosion performance with an impedance value of 16 M after 120 h immersion in a solution of 3.5 % NaCl. This attests for the passive behavior of the layer and a large charge transfer resistance.
Relying on the same methodology, Tedim et al. prepared ZnAl LDH conversion coatings on AA2024-T3 substrate. It was also the first time that a corrosion inhibitor was intercalated within the LDH galleries and its release by anion exchange with chlorides was verified [48] . After the preparation of ZnAl LDH-NO 3 , anionexchange reaction in a vanadate containing solution was performed at pH 8-9 at 50 °C > T. XRD patterns showed diffractions typically ascribed to Zn-Al LDH-NO 3 and LDH-VO x nanocontainer powders. Moreover, the obtained surface revealed an uneven film structure, with micro-sized islands separated by a thin layer covering the remaining surface. This differentiated growth was attributed to a higher dissolution of aluminum matrix at sites where intermetallic particles existed [48] . EIS measurements and optical photographs ( Fig. 15 ) showed that the protection of AA2024 substrate was significantly improved when LDHs were formed ( Fig. 15 ).
Three other studies [ 79 , 84 , 338 ] followed to provide a comprehensive understanding of the influence of different parameters on the LDH growth and the LDH layer corrosion performance. The first was in respect to the effect of AA2024-T3 surface pre-treatments on the LDH growth. Different alkaline based pre-treatments were tested including existing industrial treatments [79] . The second study aimed at the optimization of the experimental synthesis conditions (the concentration of zinc nitrate and presence of corrosion inhibitor) for the fabrication of LDH films [338] . The results from these two studies [ 79 , 338 ] revealed that a compromise must be found between the stability of the native oxide film and the LDH formation to achieve a higher corrosion resistance, while the extent of surface coverage can be controlled by the concentration of reactants added into the synthesis bath. Finally, the authors provided an understanding of the mechanism of corrosion protection of these nanocontainer layers when intercalated with nitrate and vanadate, by using EIS and SVET ( Fig. 16 ) as two main electrochemical investigation tools [84] . According to their investigation, the ZnAl LDH conversion coatings intercalated with NO 3 − , assume the role of a physical barrier against aggressive species (e.g. Cl − and O 2 ), which includes as well as nano-trapping property of LDH. An additional protection mechanism was also associated to the presence of NO 3 − and Zn 2 + (present in the LDH hydroxide layers), as both species promote corrosion inhibition effects. After intercalation with vanadate, the obtained LDH films demonstrated an even superior corrosion resistance, associated to the physical barrier nature of the LDH as well as to the capacity of the vanadate to be released and act as efficient corrosion inhibitor due to their oxidative power and formation of a protective layer at the surface, hence impeding the ORR reactions.
Several other approaches based on LDH-CCs for the corrosion protection of Al alloys were proposed since then [ 52 , 243 , 335 , 339 ]. Wang et al. described the synthesis of LDH coating by immersing Al substrates into a solution already containing pre-formed MgAl LDH particles for 36h at autoclave conditions (T = 100 °C) [335] . The obtained LDH coated samples were subsequently modified by a further immersion in a solution containing 8-HQ at RT, pH 11 for 6h. The microstructure of the overall LDH coating intercalated with 8HQ − was preserved and an efficient active anticorrosion layer was obtained. Zhang et al. [340] prepared ZnAl LDH film intercalated with vanadate, molybdate and MBT, on AA2024. Although it was stated that the respective anions were successfully intercalated between the LDH interlayers, the shift of the (003) and (006) XRD reflections was not clearly demonstrated. However, the electrochemical and SST tests showed an improved corrosion inhibition with all LDH films. The highest performance was recorded for ZnAl LDH-VO 3 and the lowest with ZnAl LDH-NO 3, accordingly with LDH-VO 3 > LDH-MBT > LDH-MoO 4 > LDH-NO 3 . The same group stated that they managed to synthesize Ce-doped ZnAl-LDH film on AA2024 through urea hydrolysis [341] . The formed LDH films were afterwards loaded with V 2 O 7 4 − . The results of EIS measurements demonstrated a good corrosion resistance of the "doubledoped" LDH films which was associated to a synergistic inhibition effect of the combination of released vanadate anions and dissolved Ce 3 + from LDH structure. Nevertheless, no clear evidence of the Ce 3 + cations insertion between the LDH galleries or within the hydroxide layers, was demonstrated. This implies that the Ce 3 + could have simply been present/adsorbed on the surface of the LDH film and contributed to the overall corrosion inhibition.
Recently, Iqbal et al. reported for the first time the possibility to produce LDH thin films with Ca 2 + and Al 3 + cations on Al alloys [316] . The CaAl LDH conversion layers intercalated with nitrate were prepared on AA6083 substrate, by a one-step hydrothermal synthesis at 140 °C (under autoclave conditions) and a pH = 10. Different immersion times were tested with t = 24 h being the best condition for an optimized corrosion resistance. Indeed, the EIS measurements showed excellent barrier properties with an impedance value at 0.01 Hz close to 3 orders of magnitude higher ( Fig. 17 ) in comparison to the bare substrate. The authors also highlighted that even enhanced corrosion resistance can be obtained with further intercalation with corrosion inhibitors.
The treatment of Al alloys with LDH-CC coatings can provide an active corrosion protection against aggressive environments and micro galvanic cells as a result of the presence of IMCs, but the role of LDH-CC can also be extended to hybrid multi-material assemblies that are responsible for accelerating corrosion as a result of a macro-galvanic effect. This possibility has been already shown by the use of LDH pigments [236] , but a recent work described a methodology of implementing ZnAl LDH conversion coating as a pre-treatment before a hybrid friction spot joining of AA2024-T3 to CF-PPS (carbon-fiber reinforced polyphenylene sulfide) [342] . In trying to minimize the galvanic corrosion effect resulting from the direct contact of AA2024-T3 with the much noble CF-PPS, the AA2024-T3 part was subjected to a pre-treatment by ZnAl LDH incorporated with nitrate and vanadate (model corrosion inhibitor). The adhesion performance of the joints after LDH treatment was tested by the lap shear method (ASTM D 3163) and revealed an improvement close to 20 %. After being exposed to SST (ASTM B117-16), the overall mechanical performance of the joints treated with ZnAl-LDH (nitrate) was still well maintained and a good corrosion resistance was noted in comparison to the reference joints (nontreated with LDH). However, the joints treated with ZnAl LDH containing vanadate exhibited an unusual behavior. Indeed, in comparison to the many reported studies [ 44 , 48 , 92 , 286 , 287 , 332 ] of the superior inhibition performance of vanadate with Al alloys, the joints treated with ZnAl LDH vanadate demonstrated lower corrosion resistance and adhesion performance after SST. The authors justified this behavior to the unstable nature of vanadate speciation when subjected to a variation of pH or oxidation conditions. The fact remains, that for the first time it has been proven that LDH-NO 3 conversion films can be good candidates to improve corrosion resistance of hybrid joints without compromising their mechanical performance.
The corrosion protective property of LDH-CC can be expressed is several forms. In the above studies LDH-CC provided an active corrosion protection to Al alloys through the triggered release of corrosion inhibitors. However, LDH-CC can also be exploited as a support for further chemical modification to obtain a hydrophobic layer which will delay the ingress of water and aqueous species hence minimizing the occurrence of corrosion. Wang et al. proposed the fabrication of MgAl LDHs film growth by urea hydrolysis on a pure Al substrate [312] . The subsequent surface modification with hydrophobic compounds (sodium oleate, sodium laurate, and sodium stearate), led to the increase of the surface hydrophobicity of LDHs, delaying the formation of biofilms involved in biocorrosion. The performance of the treated LDH films was found to be as following; LDH-CC (without treatment) < LDH-CC with oleate < LDH-CC with laurate ≈ LDH-CC with stearate.
The above strategy was also demonstrated by Zhang et al. where LiAl LDH layers were grown on an AA2198-T851 and modified by the adsorption of a surfactant anion PFDTMS (1H,1H,2H,2Hperfluorodecyltrimethoxysilane) ( Fig. 18 a). The investigation was supported by high contact angles (CA) measurements with water revealing a value of CA ~168 ° [46] ( Fig. 18 b).
The same superhydrophobic properties were reproduced by treating MgAl-and ZnAl LDH-CC with stearic acid [ 47 , 343 ], ZnAl LDH with laurate anions [313] , MgAl and ZnMgAl LDH with F-triethoxy-1H,1H,2H,2H-tridecafluoro-n-octylsilane (FAS-13) [ 38 , 344 ]. In the case of laureate anions, hydrophobization was achieved by successful intercalation of this anion into the LDH galleries whereas for the others species namely PFDTMS [46] , stearic acid [47] and (FAS-13) [ 38 , 344 ], it is believed that they were bonded onto the LDH film surface through electrostatic interaction or chemical reaction with the LDH hydroxyl group present at the surface [313] . Just recently, a new study showed the possibility to implement both hydrophobicity and active corrosion protection into an ZnAl LDH film grown on an Al ( ≥ 99.9 wt. %) substrate [345] . The ZnAl LDH film was prepared by co-precipitation and the resulting LDH coated Al panels were immersed first in a vanadate containing solution followed by a second immersion in a laurate-based solution. This resulted in a ZnAl LDH-CC cointercalated with both laurates and vanadate. The electrochemical investigation using polarization and EIS revealed a superior corrosion resistance that was attributing to the superhydrophobicity from laurate, anion-exchange reaction from LDH allowing the trapping of aggressive chlorides and release of vanadate corrosion inhibitor.
Although hydrophobization treatments may contribute greatly to enhance the corrosion protection of Al alloys, the property of hydrophobicity can be useful only if applied as a last step on a multi-level coating scheme (last interface in contact with air). This is generally achieved by means of primers or top-coats. More-over, the hydrophobic effect may be lost over long immersion times or upon mechanical defect. Therefore, the LDH aptitude of self-healing/active corrosion protection is much more relevant and should be additionally implemented.
The continuous improvement of LDH-CCs drove to the addition of an extra step, which is their implementation in frame of a multi-scheme protective system where LDH would play the role of the first "defensive line", previously occupied by CCC. Therefore, attempts have been made to cover the LDH coated Al samples with an additional polymer coating and test the resulting corrosion resistance performance. Wu et al. fabricated vanadate intercalated MgAl LDH-CC on AA2024-T3 followed by the deposition of a sol-gel layer [287] . The hybrid system exhibited good adhesion (according to GBT9286-1998 standard) performance which meets the requirement for CCC replacements. More recently, Yasakau et al. reasserted the possibility to increase anticorrosion resistance by a bilayer system LDH/sol-gel. EIS and SVET investigations on an artificially defected LDH/sol-gel coated AA2024, revealed that the mechanism of corrosion protection is based on the ion-exchange concept, where the chlorides are captured and the corrosion inhibitors (vanadate) are released to act on the defected areas [346] . Moreover, the authors reiterated that the procedure used to pre-treat the AA2024 surface does have an influence on the LDH growth, a statement that supports that of Zheludkevich et al. [79] . A lab-based surface pre-treatment was compared to that of a conventional industrial pre-treatment. Although the lat- ter tends to result in a much stable and thicker Al oxide layer, it provides less dissolved Al for LDH formation. Therefore, the labbased pre-treatment was found more favorable to the formation of a high coverage LDH film, which also means a high capacity of corrosion inhibitor storage. However, independently of the pretreatment used, substantial damage was noted on the sol-gel coating above the LDH-NO 3 CC after immersion in a 0.5 M NaCl solution ( Fig. 19 a and c). On the other hand, the LDH-V 2 O 7 demonstrated a superior compatibility with the sol-gel coating ( Fig. 19 b and d), which allowed them to act as efficient active corrosion protective systems.
As demonstrated with LDH pigments and LDH-CC on bare aluminum alloys, there are various ways to adapt and modify a system to achieve an active response. The intercalation of corrosion inhibitors into LDH-CC is generally classified as an extrinsic approach to corrosion protection [195] . Interestingly, intrinsic paths to improve corrosion resistance exploiting the properties of LDH nanocontainers are also possible. Yan et al. reported the observation of an intrinsic self-healing effect taking place by a dissolution/recrystallization of the LDH film [347] . After ZnAl LDH conversion layers were produced on AA6061 substrates (LDH-CC thickness = 8 μm), they were covered by an epoxy resin which was artificially scratched and immersed in a solution of 3.5 % NaCl. A restoration of the damaged LDH film and the recovering of the area of the scratch was witnessed after a certain period of time (thickness of the reconstructed LDH was close to 5 μm).
In a similar context, Visser et al. [ 86 , 315 , 348-357 ] proposed another original concept of self-healing protection that can also be regarded as intrinsic self-healing effect. Indeed, in an extrinsic approach, the Al substrate is already covered with the LDH-CC and is loaded with corrosion inhibitors that are released upon damage ( Fig. 20 ). In the method proposed by Visser et al. [ 86 , 315 , 348 -357 ], the LDH films are not readily fabricated on the Al substrate, but in-stead LDH film formation occurs as a targeted response to a damage caused in a polymer matrix ( Fig. 20 ).
The Li-containing coatings were subjected to artificial defects and characterized by several electrochemical [ 352 , 355 , 357 ] and surface investigation [ 351 , 356 ] techniques. The authors showed that the Li from the coating can leach into the area of defect and promote the formation of an LiAl LDH conversion layer, hence providing a passive protection to the Al alloy (see Fig. 21 ). It was stated that the mechanism of formation of these LiAl LDH layers involved three main stages; (1) thinning of the Al native oxide layer, (2) anodic dissolution and LDH film formation and finally (3) continuous dissolution and growth of the LDH film.
This proposed mechanism is similar to the explanation provided by Buchheit et al. on the first studies carried out on LDH as conversion layers [11] . This prominent self-healing behavior by Li-leaching from organic coatings, has been reproduced in several studies and applied on different Al alloys (AA2024, AA7075, AA5083, and AA6014) [315] .
Going through the various works published on LDH conversion coatings, it can be presumed that the topic remains novel and incomplete. The late accomplished studies on LDH-CC on bare Al alloys show a continuous improvement in terms of the mechanistic understanding [ 248 , 358 ] and the optimization of the LDH application [ 342 , 359 ]. A list of these works can be found in Table 5 .

LDH-based sealing of anodic layers
Anodizing methods whether conventional or plasma based, are often employed for the pre-treatment of Al alloys as base before subsequent coverage with additional organic coatings. Indeed, the resulting anodized Al is porous ( Fig. 22 a), which on one hand may enhance the adhesion to subsequent coating layers, but on the other hand may be an open access for aggressive elements into the interface of the Al alloy, leading to severe corrosion. In order to maintain the durability and prevent corrosion degradation Table 5 Recapitulative of the studies achieved on LDH-CC on Al alloys a) without any additional surface treatment b) with hydrophobic properties and c) implemented in a multiframe coating system.  20. Schematic illustration of the extrinsic and intrinsic approaches (according to [347] and [353] ) to the LDH conversion coatings.
of the anodic layers, an additional post-sealing treatment needs to be performed [ 104 , 148 , 149 , 360 ]. Hot water sealing (HWS) is one of the most common sealing methods. The anodized Al parts are immersed in a hot boiling water (or steam) to allow the formation of boehmite, filling the pores of the anodic layer ( Fig. 22 b) [361] . HWS improves the barrier properties of the anodic films but does not offer any active corrosion protection, due to the absence of corrosion inhibitors. Formally active compounds such as dichromate and nickel acetate use to be added by an impregnation process to confer an additional active protection [148] . However, due to their toxicity these were replaced by more suitable corrosion inhibitors [ 149 , 362 ] for example rare earth salts [149] ( Fig. 22 c). Indeed, these corrosion inhibitors can be added during a HWS treatment or be trapped between the anodic layer and an organic coating [361] ( Fig. 22 d).
Nevertheless, the direct addition of corrosion inhibitors into the pores of the anodic layer will lead to the manifestation of similar issues encountered with the organic coatings in Section 2.3 , namely fast consumption of the inhibitor, possible interaction with the subsequent organic coating, degradation of the inhibitor …etc. This is where the LDH sealing could play a crucial role. In the following paragraphs the various advantages offered by the LDH-CC on enhancing the corrosion resistance of anodic films are reviewed.

Conventional anodizing.
Conventional anodizing of Al alloys involves an electrochemical process where an electrical current is passed through an Al alloy sample, immersed in an acidic bath. A partial consumption of the Al alloy surface takes place and an anodic film is progressively formed toward the inner part of the metal alloy. The anodizing film generally consists of an inner barrier oxide layer (10-100 nm) and an outer porous layer, with pores that would exceed 100 μm in depth [363] . Hence, explaining why a post-sealing treatment is needed. The idea with LDH post-sealing is not only to seal these pores and sustain a barrier protection, but also to compensate the lack of active corrosion protection, by means of LDH nanocontainers loaded with corrosion inhibitors.
The earliest study on this specific topic was performed by Zhang et al. [240] . ZnAl LDH films were prepared by in-situ co-precipitation process on a porous anodic alumina/aluminum (PAO/Al) substrate. The process was followed by an anion-exchange reaction in order to intercalate laurate ions. The success of the intercalation with laurate anions was confirmed by the observation of a shift of the main LDH reflection (003) and (006) at lower angles. The modified LDH films demonstrated superhydrophobic properties which were associated with the micro-and nanoscale hierarchical structures. As stated earlier in this review, superhydrophobicity is a feature that provides corrosion protection but only in a passive manner, which means no self-healing properties are achieved.
To integrate self-healing properties into LDH films, corrosion inhibitors must be added. This is what Li et al. did in their work with vanadate loaded LDH film on anodized 2198 Al alloy [364] . The anodized samples were first subject to HWS (boiling water for 20min), followed by several hours immersion in a solution containing the precursor mixture (Zn (NO 3 ) 2 6H 2 O and NH 4 -NO 3 ) at a pH close to neutral and T == 45 °C.
The formation of the ZnAl LDH films was explained by the following equations: These equations represent the mechanism of LDH formation claimed previously by other studies [ 11 , 235 ]. The last step consisted of immersing the readily formed LDH coated samples, in to a solution containing NaVO 3 at a pH = 8.8, T = 45 °C for 2 h.
The LDH films on AA2198 demonstrated enhanced corrosion protection which could be explained by the sealing of the pores, reducing the penetration of corrosive ions; and the release of vanadate ions providing a long-lasting protection. This research group also prepared LDH film with different cations Mg 2 + , Co 2 + , Zn 2 + and Ni 2 + on the same anodized AA2198 [365] . The obtained LDH films displayed different morphologies, with ZnAl LDH and NiAl LDH showing a more compact and best adhesion performance to the anodized layer. However, after hydrophobization treatment with PFDTMS, it has been found that the superhydrophobic properties were more evident with the samples treated with the Mg-and Co-Al LDH films. This was justified by the more porous structure of these two LDH films and their capacity to trap a larger amount of air. These statement were supported by the following CA values performed after PFDTMS treatment: MgAl LDH, CA = 168.8 °; CoAl LDH, CA = 169.6 °; NiAl LDH, CA = 165.8 °and ZnAl, CA = 164.2 °. Indeed, Mg-and Co-Al LDH films show higher water contact angles in comparison to the compact Ni-and ZnAl LDH layers ( Fig. 23 ).
Kuznetsov et al. applied the concept on tartaric sulphuric anodized (TSA) AA2024 [244] ( Fig. 24 a). The bilayer system was investigated by SEM and glow discharge optical emission spectroscopy (GDOES) and the results have shown that the LDH covers the surface and fills the pores of the anodic layer. It was clearly demonstrated that the active sealing enhances the overall corrosion performance and ensures an effective healing of artificial defects, which was clearly demonstrated by salt spray test according to the ISO 9227 standard ( Fig. 24 b).
In an analogue work, Mata et al. suggested the fabrication of LiAl LDH on TSA anodized AA2024 instead of ZnAl LDH films [44] . The authors performed a screening of some important synthesis parameters such as temperature and pH to identify the most suitable conditions to grow LiAl LDH with improved structural and corrosion performance. At the end of this investigation, the authors claimed that LiAl LDH intercalated with vanadate can be prepared at low temperatures such as 55 °C and 25 °C and the resulting coating has shown efficient corrosion protection guarantied both  by a barrier and active self-healing effect. These results demonstrate the superiority of LiAl LDH coatings over the conventional hot water sealing since they can be prepared at lower temperature and still demonstrate good corrosion resistance. Liu et al. compared the performance of anodized 2Al2 alloy when treated with NiCeAland ZnCeAl LDH films [366] . It was revealed that due to their strong interlayer interaction and compact structure, NiCeAl LDH containers can host more vanadate inhibitors, hence demonstrating a higher corrosion resistance in comparison to ZnCeAl LDH. This work highlights the influence of the divalent cations on the LDH structure in promoting corrosion protection. In the same perspective, MgCeAl LDH-CC were prepared on HWS anodized AA6082 [367] . EIS measurements showed enhanced corrosion resistance with the implementation of the Ce the LDH interlayers, even after 1200h. This concept was subsequently improved with an addition superhydrophobic treatment with 1H, 1H, 2H, 2H perfluorododecyl trichlorosilane (PFDTS) [368] . In the latter study, the AA6082 was first anodized using a hydrosulfuric acid solution. The obtained anodic layer was further treated with a MgCeAl LDH-CC followed by a dipping in a PFDTS based solution. The final CA of the PFDTS treated LDH was estimated to be close to 155.6 °. The results of the electrochemical investigation revealed a long-lasting corrosion resistance that also goes beyond 1200 h.

Plasma electrolytic oxidation (PEO).
Another possible application is the formation of LDH on the surface of plasma electrolytic oxidation (PEO) coated aluminum surfaces. PEO is an advanced anodizing process where ceramic-like coatings are formed at high voltages in low-concentrated eco-friendly alkaline electrolytes with visible short-lived micro discharges over the alloy surface [369] . The oxide layers developed by PEO are hard, well-adherent to the substrate and improve both corrosion and wear properties of the Al alloys [370][371][372] . However, similarly to conventional anodizing, PEO coatings always show a porous structure that can compromise the properties of the coating. In order to address this issue, several attempts have been made including selection of optimized current/voltage regimes and post-treatments [ 371 , 373-375 ]. Nevertheless, none of these attempts promote active protection to the ceramic layers nor does it prevent the porosity. Serdechnova et al. have shown that ZnAl LDH layer in-situ grown and loaded with vanadate as corrosion inhibitor on a phosphate/silicon based-PEO covered AA2024 aluminum alloy, allows to achieve "smart" active corrosion protection ( Fig. 25 a and b) [314] . The synthesis of the LDH films was carried out by simple co-precipitation method with the same conditions employed by Kuznetsov et al. [244] . EIS and SVET results have demonstrated self-healing ability for the specimen containing vanadate inhibitor.
The possibility to use LDH nanocontainers as a post-sealing treatment has generated a lot of interest, and more studies have been performed in order to understand the factors/parameters of both the PEO and LDH preparation process that could influence the LDH growth. For instance, in an attempt to prepare NiAl LDH film on AA6061, Dou et al. [376] noticed that LDH would preferably grow on the γ -Al 2 O 3 rather than the α-Al 2 O 3 phases composing the PEO coatings. To support their argument, the authors first noted that LDH formation dominated in areas where the molten alumina had a higher cooling rate, and the latter areas are claimed to be mainly composed of the γ -Al 2 O 3 phase. Secondly, they claimed that the A1-O-A1bond energy of α-Al 2 O 3 is greater than that of γ -Al 2 O 3 which means that Al is unlikely to be dissolved from the α-Al 2 O 3 . To further prove their point, XRD and FT-  IR measurements were achieved on α-Al 2 O 3 and γ -Al 2 O 3 powders before and after they were subjected to LDH growth. The (003) and (006) XRD reflections associated with the NiAl LDH were much more intense for the LDH formed on γ -Al 2 O 3 then on α-Al 2 O 3 .
The FT-IR also showed stronger absorption peaks for NiAl-LDH on γ -Al 2 O 3 in comparison to NiAl LDH formed on α-Al 2 O 3 .
Mohedano et al. conducted a detailed investigation on the influence of PEO voltage on the LDH growth [377] . Four voltages were selected (350 V, 400 V, 450 V and 500 V) and the higher the voltage, the thicker was the resulting PEO coating. Using SEM, XRD and GDEOS, the authors demonstrated that LDH growth was very limited in the case of PEO coatings obtained at higher voltages. Although the thicker PEO coatings demonstrated better corrosion performance, they could not be supported by the active cor-rosion protection provided by vanadate loaded LDH nanocontainers. It was also argued that not only the α-Al 2 O 3 phase is unfavorable to LDH growth but γ -Al 2 O 3 phase rendered the dissolution impossible as well. In other words, no recrystallization would allow the formation of LDH [378] . However, the results of this study did not reveal a complete explanation in respect to the relation between PEO parameters and LDH growth. Therefore, the same group of researchers published another study dedicated to the understanding of the influence of the PEO phase composition, prepared on AA2024, on the in-situ LDH growth [80] . The main conclusions drawn from the investigation are the following: a) both crystalline α-Al 2 O 3 and γ -Al 2 O 3 phases are unfavorable to LDH growth, while natural amorphous Al 2 O 3 and the PEO inner layer would facilitate the LDH growth, b) the tortuosity of the PEO layer together with the accessibility to Al(OH) − 4 (see Eqs. (16 ) and (18) ) are the limiting factors for LDH formation. Therefore, for optimized results, a compromise needs to be found between these conditions. In a different study, Zhang et al. [379] questioned the influence of the elements composing the PEO coating rather than its morphology. It was found that, LDH films produced on the PEO coating formed in a NaAlO 2 electrolyte were more uniformly distributed and dense in comparison to the ones prepared on a Na 2 SiO 4 based PEO. Moreover, the removal of the Si enriched surface by polishing the PEO outer layer, also contributed positively to the LDH-CC formation. In other words, the presence of silicon impedes the nucleation and growth of LDH layers.
Recently, a systematic study was carried out by del Olmo et al. [380] . Several preparation parameters for both PEO and LDH were tested to obtain an optimized LDH treated PEO coating on AA1050-H18 substrate. The investigation showed that a continuous and well-defined LDH layer can be obtained when the thickness of PEO coatings is ~1 μm. This is due to an easy access to Al(OH) 2 + cations. Moreover, the corrosion resistance evaluation of the LDH treated PEO coatings revealed that the ZnAl LDH intercalated with NO 3 − alone, does not contribute in enhancing the corrosion protection of the PEO coated substrate. The only way to reinforce the corrosion protection, is by loading the LDH nanocontainers with efficient corrosion inhibitors such as vanadate.
The above studies [ 80 , 379 , 380 ] summarized a few conditions that need to be met in order to be able to grow LDH on a PEO coated Al alloy. However, it is important to resolve situation where LDH growth on certain PEO coatings remains challenging. A potential answer would be to find a new source for Al 3 + cations. In a very recent work, the formation of a xerogel layer was added between the steps of PEO treatment and LDH post-treatment, aiming at creating a new source for Al 3 + [20] . The PEO coating on AA2024 was prepared at constant voltage mode with 450 V, 10 min and 0.5 A, and these chosen conditions were qualified as unsuitable for LDH growth [ 80 , 377 ]. However, after the addition of the xerogel layer seconded by an LDH treatment, a homogeneous and compact layer of LDH was formed over the xerogel modified PEO coating. The xerogel formation relies on the dissolution of the metal salt dominated by hydrolysis followed by condensation reactions: -Aluminum oxide gel formation: -Hydrolysis reaction (24) Nucl eophil e ( Nu ) = NO − 3 , H 2 O After formation of the xerogel layer on the PEO coated AA2024, the ZnAl LDH-NO 3 synthesis can take place according to: Although a slight corrosion improvement was observed with the samples treated with ZnAl LDH intercalated with vanadate, the performance was not significantly better in comparison to other LDH treated coatings [ 44 , 314 , 377 ]. However, this groundwork allows to broaden the possibilities of LDH growth on different treated surfaces. All relevant studies associated with the LDH treatment of conventional anodized and PEO treated Al alloys were summarized in Table 6 .

Implementation of LDH in industry
LDH presents a number of attractive features e.g. anionexchange, insertion of various elements into the structure, calcination/rehydration, thermostability, surface modification etc., that found use in various applications such as catalysis, energy and corrosion protection, as discussed in this review paper. Another important advantage of LDH is that they are easy to manipulate, costeffective and conform to REACH regulations. These factors make their implementation into industrial applications very feasible. It is important to point out that LDH in form of slurry/powder is already being produced in large scales by several companies. For example, SASOL GmbH have a list of products such as the trademark products PLURAL® (based on MgAl LDH) which are produced through a patented [381] metal alcoholate route [382] . Their product is largely applied as a catalyst support for the heterogeneous catalysis industry. SINWON chemical Co., LTD., produced a series of non-toxic and heavy metal free LDH pigment (mainly MgAl LDHs), that are commercialized as neutralizing agents for Polyolefin, heat stabilizers for halogenated polymers (e.g. PVC), heatretaining agent for agricultural films and flame retardant [383] .
In the context of corrosion protection, LDH is still in the process of development. However, there are a few companies that have dedicated their R&D to focus in this area of expertise. For instance, KISUMA Chemicals [384] , a world leader in the production of hydrotalcites, possesses a branch specifically devoted to the development and optimization of LDH pigments for corrosion protection. They aim mostly is to improve the barrier properties of polymers and provide better encapsulation solutions [384] . Another company, SMALLMATEK Lda. [385] , produces and performs research on LDH additives mainly for the purpose of corrosion protection. Their pigments can be incorporated into coatings and integrated in a multi-level corrosion protective framework.
The latter examples are not the only ones, BASF Coating GmbH has also shown interest in the development of LDH particles for corrosion protection. The company has been collaborating with research groups and exploring ways to use these pigments, especially on automotive coatings [ 17 , 54 , 386 ]. More recently, AkzoNobel Coatings GmbH has proposed a novel alternative to chromatebased coatings that is based on lithium technology. The concept has been discussed earlier in the section of LDH as functional coatings for bare Al alloys [ 315 , 353 , 356 ]. The idea is to provide a selfhealing property to a coating by a triggered (e.g. scratch) leaching of Li salts from a polymer matrix which would react with the interface (e.g. dissolved Al) and form a LDH protective layer on the area of the damage. This patented method is commercially available for the corrosion protection of aeronautical parts [ 387 , 388 ].
Since the first reports on the use of LDH for corrosion protection [9][10][11] till recently, important progress has been made in terms of process parameters, morphology and implementation in multi-level corrosion protective systems. In this review only case studies on Al alloys have been cited, however the use of LDH has been applied successfully to various metal substrates such as zinc [ 81 , 248 ]  Sealing and active corrosion protection [20] 395 ]. This is an important feature of LDH since it could be advantageous for a wide range of industrial applications. Nevertheless, there is still room for improvement when it comes to operating conditions, such as processing time and temperature, stability in various conditions (e.g. different pH ranges) and coating compatibility and adhesion performance. These are just a few points among others to overcome, but this was the case for most innovation technologies at the first stage of its development. A summary of the LDH (in form of pigments and conversion coatings) properties and advantages is provided in Table 7 . Synthesis /processing -Possibility for preparation at 60 °C > T and reduced synthesis and ageing time.
-Requires no additional accelerators.
-Diversity of preparation methods.
-Possible formation of unwanted products but may be avoided by optimizing processing condition. -Waste highly dependent on the precursors used for LDH preparation, but overall non-toxic. (See below) -Interestingly LDH pigments are used to recycle waste and minimize sludges generated from other industrial processes.
-Unwanted products can be easily avoided.
-Waste highly dependent on the precursors but overall is minimal and non-toxic [338,359] pH stability -LDH release mechanism can be induced by anion exchange at basic medium or by dissolution in acid medium. However, the dissolution is not always desired and depends on the composition of the hydroxide layers cations and the interlayer anions. -The optimum pH range of obtaining LDH can be interpreted as the optimum domain of existence/stability of LDH, which can be summarized for various LDH types as: -pH 4.5 < LDH stability < pH 11.5 [400,401] -Similarly, to the LDH pigments, the risk of dissolution in acidic environments may compromise their corrosion protection abilities. -Not much studies have been performed in investigating the pH influence on LDH -CC performance. [402] Compatibility -This parameter can be related to the pH stability of LDH pigments and their ability to mix well with organic coatings. Several LDH surface modifications were proposed to overcome this problematic and good results were achieved. -Non-compatibility of LDH pigments may impede their good functionality.
-Compatibility also depends on the concentration used in the organic coating.
Reported concentration from 2 to 15 % of LDH pigment in polymer coatings were used [15,17,[54][55][56]286,300,[304][305][306][308][309][310] -The reported studies on subsequent addition of organic coating on LDH-CC do not show any incompatibility issues. -The surface modifications achieved on LDH-CC to obtain hydrophobic properties is an indication that this pathway can be followed to promote further compatibility with subsequent organic coatings. [287,312,343,344,346] Adhesion/risk of disbonding -The risk of disbonding due to the addition of LDH pigments into an organic coating was often pointed out. This is also related to the above compatibility factor. However, the risk has been minimized with the recent development in this field. [18,286] -The adhesion tests performed on LDH-CC with or without subsequent addition of organic coatings demonstrated reasonable results. -Moreover, as mentioned above, improving the compatibility by LDH-CC surface modification can enhance adhesion and minimize the risk of disbonding. -The combination of anodized Al alloy and LDH-CC can ensure further strengthening of the adhesion properties. [287,337,346] ( continued on next page ) Corrosion protection mechanism -Active corrosion protection/Self-healing through inhibitor release or reconstruction -Corrosion sensing -Aggressive anions trapping (eg. Cl − , SO 4 2 − ) -Some studies reported comparable to better corrosion resistance from LDH pigments in comparison to Cr(VI) pigments. [15,18,34,35,83,238,289,299] -Active corrosion protection/Self-healing through inhibitor release or reconstruction -Corrosion sensing -Aggressive anions trapping (eg. Cl − , SO 4 2 − ) -Barrier protection -Post treatment /sealing -Hydrophobicity -LDH-CC corrosion resistance performance was reported to be comparable to better than CCC. [8,9,20,46,48,80,240,244,312,313,315,338,355,356,[364][365][366][367]377,380,387] Toxicity -Some of the LDH pigments used for corrosion protection applications are also considered for drug delivery therefore they were cleared for any risk of toxicity. However, more studies should be performed in this direction. [403,404] -The toxicity data available for LDH pigments are also applicable for LDH-CC Applicability to metals -Aluminum alloys, magnesium alloys, zinc and different steels [12,13,299,389,395,[405][406][407][408][409][410][411][412][413] -Aluminum alloys, magnesium alloys, zinc and electrogalvanized stainless steel [10,48,81,[244][245][246]248,314,390,[414][415][416] Cost and availability -Depending on the precursors used to synthesis LDH pigments, they can be cost effective and widely available. A few types of LDH pigments are also readily available in nature. -The energy consumption required for LDH synthesized can be minimized by an optimization of the synthesis parameters (e.g. lowering the temperature) -Only low amounts of LDH pigments is needed to provide corrosion protection [18,54,55] -The cost-effectiveness of LDH-CC is dependent on the starting precursors and the synthesis process parameters. -Lower concentration of cation in addition to inhibitors used for LDH-CC synthesis have proven to provide efficient corrosion inhibition. -Optimized LDH synthesis parameters have also shown promising results in terms of corrosion protection. [44,338,354,355,359]

Conclusions and outlook
Throughout the different sections of this review, layered double hydroxides have shown promising results in the area of corrosion protection. Their aptitude to demonstrate sensing, nanotrapping, barrier and active corrosion protection properties make them potential alternatives to the traditional chromate-based coating. LDH can either be used as pigments or as conversion coatings for the corrosion protection of Al alloys. This means that their use can be adjusted according to the desired outcome. They can be used to compensate the lack of active corrosion protection of a barrier coating (e.g. organic coating) or serve as the first functional protective layer of an Al alloy. Indeed, remarkable results have been obtained on the latter applications with a continuous improvement.
LDH has already filled a few requirements to be considered as future replacement to Cr(VI)-based coatings. Indeed, in addition to the proven anti-corrosion performance, LDH has the advantage to be environmentally friendly, cost effective, accessible and applicable to a wide range of metal substrates. Nonetheless, LDH is not the only emerging alternative to chromate and, therefore, need to fit into other requirements such as processing parameters, compatibility with other coatings and adhesion performance. These factors can be overcome by further optimization of the LDHs both in form of pigments and conversion coatings. It is important to perform more systematic studies on environmentally friendly inhibitors that could be intercalated, since they are the main feature for active corrosion protection funtionality. These studies should include modelling approaches to build a data base that would facilitate the access to the synthesis parameters and anion-exchange, which would help to obtain the best outcome in terms of active corrosion protection. The other point is to test LDH in a multi-level protective frame. The simulation of a real multi-layer coating system would be important to anticipate issues related to adhesion/disbonding or blistering caused by lack of compatibility.

Declaration of Competing Interest
No competing interests can be declared.

Acknowledgements
This work was the result of the participation of several projects namely the European FP7 project "PROAIR" (PIAPP-GA-2013-612415), the European project MULTISURF and FUNCOAT in frame of the H2020-MSCA-RISE, grant agreement No 645676 and No 823942, respectively. M.S. and M.L.Z. are also thankful to I2B fond for financial support of this work in frame of MUFfin project as well as the ACTI-COAT project (Era.Net RUS Plus Call 2017, Project 477).
This work was also financed by Portugal 2020 through Euro-

Data availability
No datasets were generated or analyzed during the current study.