Effect of excess hydrogen bond donors on the electrode-electrolyte interface between choline chloride-ethylene glycol based solvents and copper

Journal Pre-proof Effect of excess hydrogen bond donors on the electrodeelectrolyte interface between choline chloride-ethylene glycol based solvents and copper

Jorge D. Gamarra, Kristof Marcoen, Annick Hubin, Tom Hauffman PII:

S1572-6657(19)31000-8

DOI:

https://doi.org/10.1016/j.jelechem.2019.113732

Reference:

JEAC 113732

To appear in:

Journal of Electroanalytical Chemistry

Received date:

25 July 2019

Revised date:

28 November 2019

Accepted date:

5 December 2019

Please cite this article as: J.D. Gamarra, K. Marcoen, A. Hubin, et al., Effect of excess hydrogen bond donors on the electrode-electrolyte interface between choline chlorideethylene glycol based solvents and copper, Journal of Electroanalytical Chemistry(2019), https://doi.org/10.1016/j.jelechem.2019.113732

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier.

Journal Pre-proof

Effect of excess hydrogen bond donors on the electrode-electrolyte interface between choline chloride-ethylene glycol based solvents and copper. Jorge D. Gamarraa,∗, Kristof Marcoena,∗, Annick Hubina,∗, Tom Hauffmana,∗ Research group on electrochemical and surface engineering (SURF), Materials and chemistry department, Vrije Universiteit Brussel, Pleinlaan 2, Brussels, Belgium

ro

of

a

-p

Abstract

The fine tuning of deep eutectic solvents (DES) via the addition of co-solvents

re

such as water or excess ethylene glycol, to improve the electrodeposition of

lP

metals is becoming increasingly common. However, as these changes become commonplace, our understanding of their effect must deepen as well. In this

na

work, we look into how water, a secondary hydrogen bond donor (HBD), added to choline chloride:ethylene glycol (1:2 ChCl:EG) DES changes the

ur

electrochemical behaviour of the solution. We use time-of-flight secondary ion mass spectrometry (ToF-SIMS) to identify and map the organic molecules

Jo

that have interacted with the surface. A copper surface was used as an electrochemically active substrate. The molecular maps provide insights into how solvent composition changes electrolyte reactivity, as well as showing the main driver behind the entrapment of organic compounds in metal deposits from DES. This behaviour was studied at different potentials based on the information provided by linear sweep voltammetry data of the copper in the solution. These results are then contrasted against a 1:4 ChCl:EG ∗

corresponding author

Preprint submitted to Journal of Electroanalytical Chemistry

December 8, 2019

Journal Pre-proof

solution, where the excess ethylene glycol acts as secondary HBD, in order to determine the individual contributions of the co-solvents and how they affect the electrochemistry of the media and their reactions with the copper electrode. This information can be used as a set of guidelines for the tuning and optimisation of ethylene glycol based DES for metal deposition.

of

Keywords: Time-of-flight secondary ion mass spectrometry, Non-aqueous

ro

solvents, Copper

-p

1. Introduction

re

Deep eutectic solvents (DES) are a tunable and sustainable low-cost alternative to ionic liquids [1, 2, 3] with a variety of applications in electro-

lP

chemistry such as the electrodeposition of metallic nanoparticles and coatings [4, 5].The existence of DES relies on strong interactions between their com-

na

ponents to lower the melting point and form a stable liquid phase. Typically, a quaternary ammonia salt such as choline chloride acts as a hydrogen bond

ur

acceptor (HBA), and an organic molecule like ethylene glycol (EG) as a hydrogen bond donor (HBD). DES are water stable and hydrophilic [6], which

Jo

makes their transition into industrial scale applications much smoother than ionic liquids based technologies. Water, among others, acting as a secondary HBD or co-solvent can be used to improve and fine tune the desirable properties of the DES [7, 8]. Every change in the composition of the DES increases the complexity of the behaviour of these mixtures. Thus, a fundamental understanding of the interactions involved between the tuned DES and metals is of great importance. In this work, we focus on how these added HBDs can affect the species that are formed and how they bind with the electrode

2

Journal Pre-proof

surface. Conductivity, viscosity and electrochemical window are the most commonly tuned properties in DES for electrodeposition [9, 5, 10, 11]. In order to reduce the viscosity and increase the conductivity co-solvents like water and ethylene glycol are added to the DES. The addition of 1 mol eq (6.5%w)

of

of water reduced the viscosity by 30% and increased the conductivity by 40%

ro

[12] through the disruption of the hydrogen bonds between the components. On the other hand, water can limit the electrochemical performance of the

-p

electrolyte, as hydrolysis and water-choline reactions take place on the electrified interface [13, 14, 15]. An alternative to water is the use of ethylene

re

glycol. Miller et. al.[11] improved the electrodeposition yield of iron by in-

lP

creasing the ratio of ethylene glycol to choline from 1:2 to 1:4. Although the conductivity of the 1:2 and 1:4 solutions did not increase, it is still unknown if

na

any parallels can be made between the addition of water and ethylene glycol as co-solvents and what are their individual contributions to the properties

ur

of the solvent.

The changes in the electrolyte when water is added to DES have been

Jo

approached in a multitude of ways, such as: Brillouin spectroscopy [16], Pulse-field gradient nuclear magnetic resonance (PFG-NMR) [6, 17]. Neutron diffraction [18] and contact mode atomic force microscopy [19] have been used to focus on the electrode-electrolyte interface of non electrochemically active surfaces e.g graphite, platinum. Some studies on the effect of the addition of water report the occurrence of microscopic phase segregation between aqueous and DES phases [16, 19, 14, 20]. This could mean that there is more than one electrode-electrolyte interface, leading to different reactions with

3

Journal Pre-proof

the electrode surface. On the other hand, the interactions between metal surfaces and DES, have been probed by inductively coupled plasma optical emission spectrometry (ICP-OES) and total organic carbon (TOC) for the electrolyte. The electrodes were studied using scanning electron microscopy (SEM) and energy-

of

dispersive X-ray spectroscopy (EDX)[21]. In this work, copper was chosen

ro

as model electrochemically active surface and its reactions with the segregated aqueous and DES phases can be studied through the use of time-of-

-p

flight secondary ion mass spectrometry (ToF-SIMS).The effects of DES on the morphology and the different reactivities of copper have been shown by

re

Gosh et.al and Gu. et. al.[22, 23]. In those works, the effect of solvent

lP

decomposition and the role of additives such as ethylene diamine (EDA) on the deposition of copper, were studied, but the question of the mechanism by

na

which EDA, affect the deposition, and the effect of the electrolyte decomposition on the deposits remain open. In a previous study[24], we have shown

ur

the advantages that ToF-SIMS offers to the study of the interface in DES between adsorbed species and an electrochemically inactive electrode (glassy

Jo

carbon). Tof-SIMS provides detailed mass spectra of µm areas of the top 5 nm of the surface which allows us to distinguish between multiple interfaces and any separate reactions on that happen with copper. Our work on glassy carbon showed how the presence of excess ethylene glycol changed the decomposition route and by-product accumulation of DES on an electrochemically inactive surface. In some cases, DES without excess HBD, accumulated on the surface in response to potential, inhibiting the deposition of metal and inducing corrosion on the deposit. On the other hand,

4

Journal Pre-proof

this adsorption of organics could be avoided by using larger amounts of EG. In this system, the choline is less likely to accumulate but the decomposition route changes preferring an electrochemical decomposition over a Hoffman elimination. In this work, we focus on how choline and its decomposition products interact with a copper substrate and the role of increasing amounts

of

of water on the reactivity. In order to distinguish the contributions of water,

ro

we compare the performance of the DES to a similar mixture (1:4 ChCl:EG)

-p

where excess amounts of ethylene glycol acts as secondary HBD.

re

2. Materials and Methods

Choline chloride (Sigma-Aldrich, >98%) was mixed with ethylene glycol

lP

(Merck, 99.5%) in molar ratios 1:2 and 1:4. The mixtures were stirred and kept at 80◦ C for 6 hours and nitrogen gas was bubbled through the solution

na

before the experiments. A Karl Fischer coulometer was used to control the water quantity before each experiment and an average value of (1.0 ± 0.1)

ur

weight % of water was obtained. Deionised water was added afterwards and

Jo

stirred until thoroughly mixed. The solutions are refered to as: (1:2LW) for the 1:2 ChCl:EG solution with 1% water; (1:2HW) for 1:2 ChCl:EG solution with 10% water and finally 1:4 for the 1:4 ChCl:EG solution. Copper (>99%) was used as the electrode material. The electrochemical experiments were done on a 3mm diameter disc mounted in resin for the linear sweep voltammetry and 10x35 mm plates were used for the ToF-SiMS analysis. All samples were polished with 0.05 µm alumina, dipped in pickling solution (10%HNO3 , 1% HCl) for 1 minute before deposition, and then rinsed with methanol. Linear sweep voltammetry (LSV) and chonoamperometry (CA) measurements 5

Journal Pre-proof

were performed in an electrochemical cell with a three-electrode configuration, consisting of a silver wire quasi reference electrode (Ag/AgCl QRE), a dimensionally stable counter electrode, made of titanium oxide covered with ruthenium indium titanium oxides, and a copper working electrode. All the potentials mentioned throughout the manuscript refer to Ag/AgCl QRE

of

constructed by inserting a chloridised silver wire into a fritted glass capillary

ro

filled with 1:2 molar ratio choline chloride to ethylene glycol. LSV was done between OCP and -1.15 V vs Ag/AgCl QRE at 1 mV/s for all solutions. CA

-p

measurements were done at selected potentials for 600s, in order to ensure

and methanol and air dried.

re

steady state conditions. The samples were then rinsed repeatedly with water

lP

ToF-SIMS measurements were performed with a TOF.SIMS 5 instrument from ION-TOF GmbH (M¨ unster, Germany). Positive ion mass spectra were

na

acquired using a 30 keV Bi3+ primary ion beam operated in the high current bunched mode for high mass resolution (approximately 8000 at 29 µm

ur

(29 Si+ )). In this mode, a lateral resolution of 3 µm is achieved, with a surface sensitivity of 1-5nm. The pulsed ion beam target current was approximately

Jo

0.60 pA. Large area images were acquired following the protocol described in Gamarra et. al. [24]. Deviations (either negative or positive) with an absolute value below 50 ppm are indicative of good assignments. In each spectrum or image, intensities are normalised to the total ion intensity, to reduce topographic and matrix effects. This allows for a relative quantification between samples with similar chemistries [25, 26].

6

Journal Pre-proof

3. Results 3.1. Linear Sweep voltammetry on copper electrodes in the ChCl:EG based electrolytes The electrodes were polarised from their open circuit potential (EOC )

of

to -1.2V (vsAg/AgClQRE), when the reduction of hydroxyl groups becomes the dominant reaction[27], via linear sweep voltammetry. The current density

ro

measured on the Cu surface approaches -1.0x10−3 A/cm2 (figure 1), which fits

-p

one of the definitions used for electrolyte decomposition [28]. Based on the voltammograms presented in figure 1, different polarisa-

re

tion potentials were chosen to study the interactions between electrolyte and electrode surface. First, at the open circuit potential (EOC ), any electrode-

lP

electrolyte interaction will come only from the affinity between solvent and electrode. As the potential becomes more negative, -0.60V was chosen as

na

there is a reduction process on copper for the 1:4, that is absent in 1:2 regardless of the water content. After this, a third potential was chosen at -0.95V

ur

because at this point, the 1:2HW shows the onset of a reduction reaction.

Jo

This is in accordance with the literature reports of a reduced electrochemical window in the presence of water, due to hydrolysis [29, 27]. Finally, at -1.15 V vs QRE, all the available OH- groups from water and the electrolyte are being electrochemically reduced. These electrochemical profiles provide the framework for the ToF-SIMS analysis.

7

lP

re

-p

ro

of

Journal Pre-proof

in 1:2 and 1:4 ratios.

na

Figure 1: Linear sweep voltammetries (LSV) on copper electrodes in ChCl:EG mixtures

ur

3.2. Study of the potential dependent interactions between Copper electrodes

Jo

and ChCl:EG Solutions Copper electrodes were immersed in ChCl:EG mixtures for 600s under varying polarisation potentials i.e., open circuit potential (E(OC) ), -0.60V, -0.95V and -1.15V. The choice of fragments and common contaminants on ToF-SIMs spectra are treated the same way as in previous work [24]. Figure 2 shows an overlay of ToF-SIMS mass spectra acquired from large area (1 mm x 2.5 mm) imaging on an exposed copper substrate at (E(OC) ). Sodium and (SiC3 H9+ ) polydimethylsiloxane are common contaminants, but do not overlap with any fragment of interest, and therefore are of no concern. 8

na

lP

re

-p

ro

of

Journal Pre-proof

Figure 2: Overlay of ToF-SIMS mass spectra a) (0-109 m/z) and b(110-200 m/z) obtained

ur

on copper substrates exposed to a 1:2 ChCl:EG mixture and a 1:4 ChCl:EG solution

Jo

This analysis is divided in 2 sections. The first section focuses on the copper containing compounds found on the surface. These will be used to determine how an electrochemically active surface interacts with the different electrolytes. The second section focuses on the changes in the molecular fingerprint of the electrolyte, which covers choline and ethylene glycol derivatives, to provide information on possible interactions not involving ionic interactions with the surface.

9

Journal Pre-proof

3.2.1. Effect of the electrolyte on the electrode surface ToF-SIMS data analysis focused on copper containing compounds was performed, in order to evaluate the effect of the DES on the working electrode surface. Table 1 shows the most relevant fragments for this analysis. First, the presence of copper molecules proves that the surface of the elec-

of

trode is being probed. Cu+ , with a nominal mass of 62 m/z, is shown as

ro

a reference. Next, CuC2 H4 O2+ , with a nominal mass of 122 m/z, is representative for ethylene glycol-copper complexes formed during the corrosion

-p

of copper by ethylene glycol in the presence of chlorides [30]. Additionally, the common corrosion product Cu2 OH+ , with a nominal mass of 142 m/z

re

was also recorded. Finally, copper bonded to choline and its byproducts was

lP

found in the form of C5 H13 N OCu+ and C5 H11 N Cu+ , with nominal masses of 166 and 148m/z respectively. The proposed structures for these fragments

Mass (amu)

Deviation (ppm)

Resolution

62.9312

26.4

5338

CuC2 H4 O2+

122.9550

39.4

4579

Cu2 OH +

142.8618

2.9

6444

C5 H11 N Cu+

148.0183

0.4

6558

C5 H13 N OCu+

166.0262

26.4

5338

ur

Fragment

na

are presented in figure 4.

Jo

Cu+

Table 1: Mass fragments associated to copper containing fragments on copper electrodes

10

Journal Pre-proof

of

Figure 3: Proposed structures for Copper complexes

ro

Cu+ and Cu2 OH+ are evenly distributed and their relative presence does not

-p

seem to be affected by either the composition or the potentials applied. The large area images for the fragments of interest are presented in the support-

re

ing information. On the other hand, the copper complex (C5 H13 N OCu+ )

lP

shows a potential dependent behaviour, especially upon the addition of HBDs

Jo

ur

na

(figure 4).

11

lP

re

-p

ro

of

Journal Pre-proof

Figure 4: Relative intensities of C5 H13 N OCu+ a) 1:2 ChCl:EG LW vs 1:2 ChCl:EG HW c) 1:2 ChCl:EG HW vs 1:4 ChCl:EG; and C5 H11 N Cu+ b) 1:2 ChCl:EG LW vs 1:2 ChCl:EG

ur

potentials.

na

HW d) 1:2 ChCl:EG HW vs 1:4ChCl:EG on copper electrodes under varying polarisation

Figure 4a shows how the increase in C5 H13 N OCu+ , in 1:2HW, follows

Jo

the increase in current seen in figure 1 at -0.95V described in the previous section. In the absence of water, 1:2LW, C5 H13 N OCu+ is independent of potential and is distributed over the surface. Nevertheless, when water is added to the solvent (1:2HW) the amount of C5 H13 N OCu+ becomes dependent on the potential. There is a sudden increase of C5 H13 N OCu+ by one order of magnitude when -0.95V is applied, followed by an equal drop when -1.15V is applied. This is caused by water displacing choline at the more negative potential as illustrated in figure 5. This displacement of choline

12

Journal Pre-proof

by water at electrified interfaces has been modelled recently by Mamme et. al.[14]. Once choline has been displaced, the breakdown of water further displaces the organic residues through the formation of H2 . Meanwhile, figure 4c shows the behaviour C5 H13 N OCu+ in 1:2HW and 1:4. Here, 1:2HW has a higher concentration of C5 H13 N OCu+ at -0.95V, the potential where the

of

electrolyte decomposition begins, which suggests that this compound is an

ro

intermediary step in the decomposition of DES. On the other hand, the 1:4 solution shows a maximum of choline adsorption at -0.60V, the same po-

-p

tential where the limiting current is observed, which matches the data for C5 H13 N OCu+ . Based on previous studies on the orientation and decom-

re

position of choline during deposition [27, 31], we show that the passivation

lP

of the surface described in them, can be ascribed to not only the formation of choline hydroxide but to the formation of choline metal oxides as well.

na

This trend could seriously limit the application of DES for the deposition of metals that form stable hydroxides[32].

ur

Figures 4b and d show the behaviour for C5 H11 N Cu+ , a minor product of the dehydration of choline. C5 H11 N Cu+ in 1:2 LW and 1:2 HW(figure

Jo

4b) appears to be independent from both potential and composition. Finally in figure 4d, the effect of excess ethylene glycol is seen, where the amount of C5 H11 N Cu+ diminishes proportionally to the potential applied. We cannot at the moment suggest a reaction for the further decomposition of C5 H11 N Cu+ . Ethylene glycol reacts with copper forming CuC2 H4 O2+ . This is consistent with the formation of similar bidentate complexes observed on iron [24]. CuC2 H4 O2+ , is present at much higher proportion than free ethylene gly-

13

Journal Pre-proof

col. In the work of Gu.et.al [23], when ethylene diamine (EDA) is added to ChCl:EG DES the reduction peak shifts to more negative potential, where the EDA is inhibiting the reduction of EG. The observed bidentate complexes are not exclusive to ethylene glycol.As proposed by Mamme. et. al[33] the urea in ChCl:Urea DES reorients itself at the interface so that the amide

of

group is closer to the interface when negative potentials are applied. The

ro

mechanism for the formation of these bidentate metal complexes remains to be determined, and the difference in behaviour of N-groups and O-groups

-p

needs to be studied further (figure 5).

On the other hand, the formation of the described choline fragments

re

poses an interesting situation. Choline’s charged nitrogen is the part of the

lP

molecule that interacts first with the electrode surface according to electrostatic attraction (figure 5) but it is the tail end OH that reacts with

Jo

ur

C5 H11 N Cu+ .

na

copper, leading to the dehydration of the side chain and the formation of

Figure 5: Possible choline adsorbates/intermediaries on copper electrodes.

14

Journal Pre-proof

3.2.2. Effect of the applied potential on the decomposition of the electrolyte In this section, the focus is on the SIMS fragments associated to the solvent. Ethylene glycol can be identified by its molecular fragment C2 H5 O2+ with a nominal mass of 61 m/z, but its breakdown fragments are not easily distinguishable from the carbonaceous contamination that is always present

of

on any kind of sample exposed to the atmosphere. On the other hand, choline,

ro

which contains nitrogen, can be easily recognised. Fragments associated to a choline residue were found in the mass spectra. The most relevant ones are

-p

listed in table 2. A peak found at a nominal mass of 104 m/z was assigned to C5 H14 N O+ and represents the molecular fragment of choline. Table 2 shows

Fragment

Mass (amu)

Deviation (ppm)

Resolution

C5 H14 N O+

104.1097

26.4

5338

C2 H5 O2+

61.0303

5708

C3 H8 N +

na

31.1

58.0680

49.7

4657

C3 H9 N +

ur

lP

re

the mass and resolution of the fragments of interest from the electrolyte.

59.0741

19.4

4565

86.1006

48.1

4603

Jo

C5 H12 N +

Table 2: Mass fragments associated to a choline and ethylene glycol residue on copper electrodes

The behaviour of the molecular fragment, C5 H14 N O+ , as a function of potential and composition is analysed in Figure 6. First, at E(OC) , 1:2LW and 1:4 show much larger amounts of C5 H14 N O+ adsorbed on the surface than 1:2HW. At this point, the affinity between electrolyte and electrode depends 15

Journal Pre-proof

on the solvation and physical properties, such as viscosity. In 1:2LW and 1:4 solutions EG is the main HBD, while in 1:2HW water plays a significant role on the solvation [19, 34]. At -0.60V 1:4 has a significant increase in adsorbed C5 H14 N O+ when compared to the other solvents. This corresponds to the limiting current seen in the LSV (figure 1). Since 1:4 is not an eutec-

of

tic mixture [11, 13], the hydrogen bonding with ethylene glycol doesn’t fully

ro

stabilise choline, and the electrochemical decomposition of choline via free radical is ongoing until higher potentials are applied and the decomposition

-p

pathway involving the tail OH group takes over. The amount of C5 H14 N O+

Jo

ur

na

lP

re

residue in 1:4 is the lowest for -0.95V and -1.15V

Figure 6: Relative intensities of C5 H14 N O+ as function of potential and composition

16

Journal Pre-proof

Choline is adsorbed on the surface (figure 6) for the 1:2 HW solutions when the potential applied reach -0.95V [33, 19, 18, 14, 27] as the OH group in choline reacts with the surface. Finally at -1.15V, the intensity of C5 H14 N O+ drops one order of magnitude in the 1:2HW. This behaviour is consistent with choline hydroxide breakdown caused by the OH− formed by the hydrolysis

of

reaction inside the 1:2HW and displacement of choline by water [13, 14].

ro

In the 1:2LW there is a smaller change in the accumulation of C5 H14 N O+ residue at hydrolysis potentials. The residual water is not enough to displace

-p

the choline product, and a inhibiting layer of choline products is formed. The electrochemical behaviours of 1:2LW and 1:2HW are almost identical

re

between -0.95V and -1.15 but have different mechanisms.

lP

On the other hand, C5 H12 N + , which appeared only as ionisation fragment of C5 H14 N O+ in previous work [24], increases significantly when water is

na

added to the DES. C5 H12 N + is independent from potential and C5 H14 N O+ in the presence of copper. Figure 7 shows the relation between C5 H14 N O+

ur

(figure 7a) and C5 H12 N + (7b) in 1:2LW and 1:2HW. 7b shows how the C5 H12 N + fragment is independent of potential in the presence of water. A

Jo

similar behaviour can be seen when comparing 1:2HW to 1:4. In figure7c, the presence of molecular fragment C5 H14 N O+ changes are dependent on the potential applied, but when compared to figure7d, C5 H12 N + is independent of potential and one order of magnitude higher than the residue found in 1:2LW. This means that the solvation of choline by the excess HBD, be it water or EG, affects the stability of choline and its decomposition route.

17

lP

re

-p

ro

of

Journal Pre-proof

Figure 7: Relative intensities of C5 H14 N O+ a) 1:2 ChCl:EG LW vs 1:2 ChCl:EG HW c) 1:2 ChCl:EG HW vs 1:4ChCl:EG: as well as C5 H12 N + b) 1:2 ChCl:EG LW vs 1:2

ur

polarisation potentials.

na

ChCl:EG HW d) 1:2 ChCl:EG HW vs 1:4ChCl:EG on copper electrodes under varying

The current reported mechanism for the decomposition of choline ions is

Jo

based on their behaviour in aqueous media [13]. Choline hydroxide is formed in the presence of OH− and then is hydrolysed through a Hoffman elimination, as seen in figure 8. In the Hoffman elimination of choline hydroxide, trimethylamine is the most stable product because it is the least substituted amine [13]. Nevertheless, in this work, the secondary HBDs increase the stability of C5 H12 N + as final product of the elimination. Figure 8 shows the possible reactions present starting at the formation of choline base (figure 8(1)), followed by the dominant route in the Hoffman elimination (figure

18

Journal Pre-proof

8(2)) and finally the minority, or least stable, product of the same reaction (figure 8(3)). The mechanism by which the minority product becomes more stable is still unclear, but based on ethylene glycol and water having the same effect, we propose that it relies on the complexation possible through the excess hydrogen bonds. Regardless of the final product, this information shows

of

that the direct electrochemical decomposition of choline plays a larger role in

ro

the breakdown of DES than what had been previously reported by Haerens et.al[13]. The increased stability seen in DES baths with high concentration

-p

of copper ions [22], can then be explained by a competition between choline

Jo

ur

na

lP

re

and the copper ions for electrons, that favors the metallic ions.

Figure 8: Choline decomposition routes

19

Journal Pre-proof

4. Conclusions In this work, we have uncovered the effects of excess hydrogen bond donors on the breakdown of choline and their effect on the the interactions with a copper surface. Using ToF-SIMS, we were able to determine the effects of potential on the solvent electrode interactions as presented in figure

of

9. We have determined that the use excess HBD, whether is water or ethy-

ro

lene glycol, in DES leads to the stabilisation of C5 H12 N + . This fragment could be the minority product from the Hoffman elimination or the product

-p

of an electrochemical decomposition of choline that starts in the hydroxyl

re

end of the choline. This process is independent from the potential applied to the copper electrode. Additionally, we describe the formation of copper con-

lP

taining compounds (C5 H13 N OCu+ and C5 H12 N + ) as intermediaries in the decomposition of choline in DES with excess hydrogen bond donors. Copper

na

containing intermediates are formed partially through the tail OH− and reach a maximum at different potentials depending on the HBD used. In the case

ur

of water, -0.95V is the potential at which there is maximum accumulation

Jo

of the residue before water displaces the choline. On the other hand, when ethylene glycol is used the maximum residue is reached at -0.60V, which means that choline can rearrange itself more freely in non eutectic ethylene glycol solutions.

20

lP

re

-p

ro

of

Journal Pre-proof

Figure 9: Observed compounds on a metallic surface exposed to choline based solvents

na

with different co-solvents

The exact role and effect of these reactions on an electrodeposition process

ur

remain to be determined. Many factors play a role during deposition such as the different reactivity of the substrate and deposit with the solvent and the

Jo

fact that their contributions to the inhibition process can only be determined on pair by pair basis. Additionally, the orientation of the molecules to the surface and their reactions depend on the DES and co-solvent used during the process. [21, 33]. Future work should focus on the influence of metal ion concentration on the residue formation process, as well as the effect of replacing ethylene glycol, for either ethylene diamine or urea.

21

Journal Pre-proof

5. Acknowledgements J.D.G acknowledges financial support by the Colombian Administrative Department of Science, Technology and Innovation (Colciencias). SURF acknowledges the Hercules program under grant agreement ZW13 07 for the ToF-SIMS measurements. T.H. acknowledges financial support by Research

ro

of

Foundation Flanders (FWO).

-p

References

[1] E. L. Smith, A. P. Abbott, K. S. Ryder, Deep Eutectic Solvents (DESs)

re

and Their Applications, Chemical Reviews 114 (2014) 11060–11082.

lP

[2] A. Li, Y. Chen, K. Zhuo, C. Wang, C. Wang, J. Wang, Facile and shapecontrolled electrochemical synthesis of gold nanocrystals by changing

na

water contents in deep eutectic solvents and their electrocatalytic activity, RSC Adv. 6 (2016) 8786–8790.

ur

[3] E. A. M. Cherigui, K. Sentosun, M. H. Mamme, M. Lukaczynska, H. Ter-

Jo

ryn, S. Bals, J. Ustarroz, On the Control and Effect of Water Content during the Electrodeposition of Ni Nanostructures from Deep Eutectic Solvents, The Journal of Physical Chemistry C 122 (2018) 23129–23142. [4] A. Abo-Hamad, M. Hayyan, M. A. AlSaadi, M. A. Hashim, Potential applications of deep eutectic solvents in nanotechnology, Chemical Engineering Journal 273 (2015) 551–567. [5] E. A. M. Cherigui, K. Sentosun, M. H. Mamme, M. Lukaczynska, H. Terryn, S. Bals, J. Ustarroz, On the Control and Effect of Water Content 22

Journal Pre-proof

during the Electrodeposition of Ni Nanostructures from Deep Eutectic Solvents, The Journal of Physical Chemistry C 122 (2018) 23129–23142. [6] C. D’Agostino, L. F. Gladden, M. D. Mantle, A. P. Abbott, I. Ahmed, Essa, A. Y. M. Al-Murshedi, R. C. Harris, Molecular and ionic diffusion in aqueous âĂŞ deep eutectic solvent mixtures: probing inter-molecular

of

interactions using PFG NMR, Phys. Chem. Chem. Phys. 17 (2015)

ro

15297–15304.

-p

[7] D. Patel, R. More, A. Parmar, K. Kuperkar, New insight into experimental and computational studies of Choline chloride-based green ternary

re

deep eutectic solvent ( TDES ), Journal of Molecular Structure 1181

lP

(2019) 295–299.

[8] A. P. Abbott, J. C. Barron, G. Frisch, K. S. Ryder, a. F. Silva, The

na

effect of additives on zinc electrodeposition from deep eutectic solvents,

ur

Electrochimica Acta 56 (2011) 5272–5279. [9] M. Lukaczynska, E. Amine, M. Cherigui, A. Ceglia, K. V. D. Bergh,

Jo

J. D. Strycker, H. Terryn, J. Ustarroz, Influence of water content and applied potential on the electrodeposition of Ni coatings from deep eutectic solvents, Electrochimica Acta 319 (2019) 690–704. [10] K. Range, V. Agieienko, R. Buchner, Densities , Viscosities , and Electrical Conductivities of Pure Anhydrous Reline and Its Mixtures with Water in the Temperature, Journal of Chemical & Engineering Data (2019).

23

Journal Pre-proof

[11] M. A. Miller, J. S. Wainright, R. F. Savinell, Iron Electrodeposition in Ionic-Liquid Electrolytes for Flow Batteries, Journal of The Electrochemical Society 164 (2017) 1–35. [12] L. Lomba, M. Artal, C. Lafuente, B. Giner, D. Lape, Thermophysical characterization of the deep eutectic solvent choline chloride : ethylene

of

glycol and one of its mixtures with water, Fluid Phase Equilibria 492

ro

(2019) 1–9.

-p

[13] K. Haerens, E. Matthijs, K. Binnemans, B. Van der Bruggen, Electrochemical decomposition of choline chloride based ionic liquid analogues,

re

Green Chemistry 11 (2009) 1357.

lP

[14] M. H. Mamme, S. L. Moors, E. A. M. Cherigui, H. Terryn, J. Deconinck, J. Ustarroz, F. De Proft, Water distribution at the electrified interface

na

of deep eutectic solvent, Nanoscale Adv. (2019) –.

ur

[15] P. Valverde, T. Green, S. Roy, Copper electrodeposition from a watercontaining choline chloride based deep eutectic solvent, ECS Transac-

Jo

tions 77 (2017) 859–864. [16] E. Posada, N. López-Salas, R. J. Jiménez Riobóo, M. L. Ferrer, M. C. Gutiérrez, F. del Monte, Reline aqueous solutions behaving as liquid mixtures of H-bonded co-solvents: microphase segregation and formation of co-continuous structures as indicated by Brillouin and 1 H NMR spectroscopies, Phys. Chem. Chem. Phys. 19 (2017) 17103–17110. [17] C. D’Agostino, L. F. Gladden, M. D. Mantle, R. C. Harris, A. P. Abbott, Molecular motion and ion diffusion in choline chloride based deep 24

Journal Pre-proof

eutectic solvents studied by H pulsed field gradient NMR spectroscopy, Physical Chemistry Chemical Physics (2011) 21383–21391. [18] O. S. Hammond, D. T. Bowron, K. J. Edler, The Effect of Water upon Deep Eutectic Solvent Nanostructure: An Unusual Transition from Ionic Mixture to Aqueous Solution, Angewandte Chemie - International Edi-

of

tion 56 (2017) 9782–9785.

ro

[19] O. S. Hammond, H. LI, C. Westermann, F. Endres, A. Y. Al-Murshedi,

-p

A. P. Abbott, G. Warr, K. J. Edler, R. Atkin, Nanostructure of the deep eutectic solvent / platinum electrode interface as a function of potential

re

and water content, Nanoscale Horizons (2018).

lP

[20] X. Meng, K. Ballerat-Busserolles, P. Husson, J.-M. Andanson, Impact of water on the melting temperature of urea+ choline chloride deep

na

eutectic solvent, New Journal of Chemistry 40 (2016) 4492–4499.

ur

[21] D. Di Marino, M. Shalaby, S. Kriescher, M. Wessling, Corrosion of metal electrodes in deep eutectic solvents, Electrochemistry Communications

Jo

90 (2018) 101–105.

[22] S. Ghosh, S. Roy, Electrochemical copper deposition from an ethalinecucl2Âů2h2o des, Surface and Coatings Technology 238 (2014) 165 – 173. [23] C. Gu, Y. You, X. Wang, J. Tu, Electrodeposition, structural, and corrosion properties of cu films from a stable deep eutectics system with additive of ethylene diamine, Surface and Coatings Technology 209 (2012) 117 – 123. 25

Journal Pre-proof

[24] J. D. Gamarra, K. Marcoen, A. Hubin, T. Hauffman,

Electrode-

electrolyte interactions in choline chloride ethylene glycol based solvents and their effect on the electrodeposition of iron, Electrochimica Acta 312 (2019) 303–312. [25] J. C. Vickerman, D. Briggs, ToF-SIMS: materials analysis by mass spec-

of

trometry, im publications, 2013.

ro

[26] Y.-P. Kim, M.-Y. Hong, H. K. Shon, D. W. Moon, H.-S. Kim, T. G.

-p

Lee, Quantitative tof-sims study of surface-immobilized streptavidin,

re

Applied surface science 252 (2006) 6801–6804. [27] L. Vieira, R. Schennach, B. Gollas, The effect of the electrode material

lP

on the electrodeposition of zinc from deep eutectic solvents, Electrochimica Acta 197 (2016) 344 – 352.

na

[28] Q. Li, J. Jiang, G. Li, W. Zhao, X. Zhao, T. Mu, The electrochemical stability of ionic liquids and deep eutectic solvents, Science China

ur

Chemistry (2016).

Jo

[29] C. Du, B. Zhao, X.-B. Chen, N. Birbilis, H. Yang, Effect of water presence on choline chloride-2urea ionic liquid and coating platings from the hydrated ionic liquid, Scientific Reports 6 (2016) 29225. [30] P. M. May, I. M. Ritchie, E. T. Tan, The corrosion of copper in ethylene glycol-water mixtures containing chloride ions, Journal of Applied Electrochemistry 21 (1991) 358–364. [31] M. Bučko, S. Roy, P. Valverde-Armas, A. Onjia, A. Bastos, J. Bajat, Voltammetric response of water in deep eutectic solvent based on choline 26

Journal Pre-proof

chloride and urea, Journal of The Electrochemical Society 165 (2018) H1059–H1065. [32] M. Bučko, M. V. Tomić, M. Maksimović, J. B. Bajat, The importance of using hydrogen evolution inhibitor during the zn and zn-mn electrodeposition from ethaline, Journal of the Serbian Chemical Society

of

(2019).

ro

[33] M. H. Mamme, S. L. C. Moors, H. A. Terryn, J. Deconinck, J. Ustarroz,

-p

F. De Proft, Atomistic Insights Into the Electrochemical Double Layer of Choline Chloride-Urea Deep Eutectic Solvent: A Clustered Interfacial

re

Structuring, The Journal of Physical Chemistry Letters 9 (2018) 6296–

lP

6306.

[34] P. J. Smith, C. B. Arroyo, F. L. Hernandez, J. C. Goeltz, Ternary deep

na

eutectic solvent behavior of water and urea-choline chloride mixtures,

Jo

ur

Journal of Physical Chemistry B (2019).

27

Journal Pre-proof

Highlights • Water and ethylene glycol affect the decomposition of choline in DES. • These cosolvents favour the formation of the minority product.

of

• ToF-SiMS shows the metallic intermediates in the decomposition.

Jo

ur

na

lP

re

-p

ro

• Metal glycolate forms independent from potential.

28

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9