Findings

Heritage metals are often coated as part of a conservation treatment to protect the metal surface from atmospheric deteriogens [1–5]. The coating of lead by long chain carboxylates by immersion has been studied extensively [6–12]. In this journal, a relatively simple deposition process has previously been shown to produce layers of lead carboxylate on polished lead which are protective against a range of environments [10]. However, heritage lead is often not polished prior to coating [13]. Therefore, this work aims to show the effectiveness of the same ethanolic-deposited coatings on a more apt sample for the context of conservation practice-corroded lead.

The samples were prepared in such a way as to simulate the corrosion of lead due to oak-emitted volatile organic compounds (VOCs) when displayed or stored in an oak display cabinet. The creation of an oak environment has been described elsewhere [11]. Polished lead coupons were enclosed within the oak environment at 50 % RH for 9 months in order to develop a measurable layer of corrosion products [14]. After corrosion, coupons were coated in the same manner as previously described [10]. XRD and EIS measurements were performed using the same conditions as described in Ref. [10].

X-ray diffraction

In Fig. 1a–c, crystalline corrosion products are detected, namely lead formate (ICDD No. 014-0825) and hydrocerussite (ICDD No. 13-0131). Lead formate, a product of formic acid [15] reacting directly with surface lead oxide, is an important intermediate to the formation of lead carbonates such as hydrocerussite. The presence of both species shows that the lead surface is still corroding actively after 9 months [11, 14].

Fig. 1
figure 1

XRD diffraction patterns for C14 (a) and C18 (b) coated corroded lead compared with the corroded lead precursor (c). All peaks are indexed apart from unidentified peaks (asterisk): lead acetate (filled circle), lead formate (filled triangle), hydrocerussite (filled inverted triangle), lead (filled square), lead ditetradecanoate (filled diamond) and lead dioctadecanoate ( diamond)

Full size image

Comparing Fig. 1c to a and b we can see that the height of the lead and hydrocerussite peaks increases by one order of magnitude on coating. This could be due to the coating formation reaction: amorphous corrosion products lead formate and lead acetate dissolve in ethanol [16] and lead carboxylates form simultaneously (as described in [10]); the atomic density of lead in formates/acetates is much higher than in the longer chain carboxylates, so the removal of the former accompanied by the growth of the latter leaves a surface layer more transparent to x-rays even if it is thicker. In addition, the general increase in peak to background ratio across the whole of Fig. 1a and b compared to c suggests that a transfer from amorphous to crystalline corrosion products [17] occurs at the same time due to recrystallization.

Electrochemical impedance spectroscopy of corroded and polished coated samples

Figure 2a shows the Bode plots from the polished and corroded coated samples. The difference in impedance between the corroded and polished coated samples is considerable. The data at low frequencies show an 83 % increase in coating impedance for lead dioctadecanoate [Pb(C18)2] and 79 % increase for Pb(C14)2 corroded samples compared to polished samples.

Fig. 2
figure 2

Impedance (a) and phase (b) plots of Pb(C14)2 (squares) and Pb(C18)2 (triangles) coatings on polished (black) and corroded (red) lead substrates. Bare lead (open circles) and corroded lead (blue circles) are also shown

Full size image

The impedance of the polished coated samples is larger than bare lead, but not larger than corroded lead. As shown in the XRD results, there is a significant layer of corrosion products on the surface of the corroded coupon which provides a protective layer against the electrolyte compared to bare lead. However by coating the samples, the improved impedance effect is not simply additive (polished coated sample plus corroded sample). This implies a significant improvement of the coating properties by forming the coating on a corroded sample. That said, the low coating resistance (5 kΩ cm2) does not represent a reasonably protective coating [1]; lead carboxylates deposited in this way merely passivate the surface and provide porous or patchy coverage.

Figure 2b shows the corresponding Bode phase plots. For polished, corroded and bare samples, two maxima (time constants) are shown showing the freely corroding nature of the metal substrate by the corrosive electrolyte. For the polished samples, the first maximum at ~70 Hz is the same for both samples. This maximum corresponds to the double layer and surface oxide and demonstrates the reproducibility of the polishing technique. The second maximum at ~10,000 Hz corresponds to the coating. For the Pb(C18)2 coating this maximum is approximately 10° higher due to the improved corrosion resistance provided by the longer carbon chain. The corroded coated samples show a time-constant (C18 shows a double maximum) at ~10,000 Hz. This is due to the combined corrosion and coating layer providing greater corrosion resistance from the electrolyte. However, the improved capacitive properties compared to the polished sample could be due to coating deposition on the corroded surface, as observed in the impedance plots.

Conclusions

By coating a corroded sample instead of a polished surface, the effects of real artefact conservation by immersion in ethanolic carboxylate solution were observed. This improved coating was demonstrated in impedance and phase plots, where the effectiveness of the coating on corroded coupons was approximately 80 % higher than polished and corroded coupons. These results are encouraging for future trials with real lead artefacts as we have demonstrated the benefit of protecting unpolished, corroded lead using lead carboxylates. However, it must be stressed that lead carboxylates are merely passivating agents for the metallic surface and comparison of the impedance data with earlier work [12] shows that a considerable increase in coverage for the non-conducting carboxylate should be possible. Nevertheless, their user-friendliness makes them accessible to conservators as a low-cost method for short-term protection of lead artefacts.