Self-Assembled Monolayer Enables Slurry-Coating of Li Anode - ACS

Feb 12, 2019 - (d) SEM and (e) phosphor elemental mapping of a focus-ion beam cut .... As shown in Figure 4a, an OPA–Li–CNT anode has been ...
0 downloads 0 Views 7MB Size
Download PDF
This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Research Article Cite This: ACS Cent. Sci. XXXX, XXX, XXX−XXX

http://pubs.acs.org/journal/acscii

Self-Assembled Monolayer Enables Slurry-Coating of Li Anode Tuo Kang,† Yalong Wang,‡ Feng Guo,§ Chenghao Liu,‡ Jianghui Zhao,§ Jin Yang,§ Hongzhen Lin,§ Yejun Qiu,*,† Yanbin Shen,*,§ Wei Lu,§ and Liwei Chen*,§ †

Shenzhen Engineering Lab of Flexible Transparent Conductive Films, Department of Materials Science and Engineering, Harbin Institute of Technology, Shenzhen 518055, China ‡ China Energy Lithium Co., No. 100, The Ninth Avenue of Xinye, West TEDA, Tianjin 300462, China § i-Lab, CAS Center for Excellence in Nanoscience, Suzhou Institute of Nano-Tech and Nano-Bionics, Chinese Academy of Sciences, Suzhou 215123, China

ACS Cent. Sci. Downloaded from pubs.acs.org by 5.101.222.81 on 02/12/19. For personal use only.

S Supporting Information *

ABSTRACT: Li metal has long been considered as the ultimate anodic material for high-energy-density batteries. Protection of Li metal in electrochemical cycling and in the manufacturing environment is critical for practical applications. Here, we present the passivation of the Li metal− carbon nanotube (CNT) composite with molecular self-assembly of a longchain aliphatic phosphonic acid. The dynamics of the self-assembly process is investigated with sum-frequency generation spectroscopy (SFG). The aliphatic phosphonic acid molecules self-assemble on the Li metal surface via the lithium phosphate bonding, while the well-aligned long chains of the molecules help to prevent corrosion of lithium by oxygen and water in the air. As a result, the self-assembled monolayer (SAM) passivated Li−CNT composite displays excellent stability in dry or even humid air, and could be slurry-coated with organic solvents. The resulting slurry-coated Li anode exhibits a high Coulombic efficiency of 98.8% under a 33% depth of discharge (DOD) at a 1C rate in full battery cycling. The concept of molecular self-assembly on Li metal and the stability of the resulting SAM layer open vast possibilities of designed reagents for surface passivation of Li, which may eventually pave the way for practical application of Li metal in secondary batteries.

■

demonstrated in suppressing Li dendrite growth,2,7 Coulombic efficiency (CE) still needs to be improved to a satisfactory level to ensure long cycle life of the battery. Furthermore, the processing of the soft, sticky, and readily corroded thin Li foil is also a formidable challenge. In this regard, a slurry-coated Li anode would be ideal because it would be fully compatible with the processes in current battery manufacturing. However, slurry-coating of the highly reactive Li metal constitutes a serious problem. In addition to the electrochemical requirements for a highperformance Li anode such as high CE, low impedance, and suppression of Li dendrites at high current density, the slurrycoating process demands the material size to be on the order of a few micrometers or smaller, and at the same time, the material must be stable toward the environment and solvent.34 Typically, Li metal is commercially available in formats of foils of about 50−500 μm in thickness25,35 or particles of 30−70 μm in diameter.34 Small Li particles would have significantly greater specific surface area and thus would be much more reactive toward the environment and the solvent. Therefore, it is imperative to stabilize micrometer-sized Li particles for

INTRODUCTION The research on Li anodes has been revived in recent years because of the ever increasing demand for higher-energydensity batteries in personal electronics and electric vehicle applications.1−3 Although considered the “Holy Grail” of energy storage materials for extremely high specific capacity (3860 mA h/g) and the lowest redox potential (−3.040 V versus the standard hydrogen electrode), the Li anode is nevertheless not applicable yet in practical systems because of prohibiting problems including growth of Li dendrites and repeated reaction with electrolytes in electrochemical cycling.3−7 Extensive research has been conducted to overcome these drawbacks of the Li anode. Several strategies such as electrolyte additives,5,8−11 3D current collector,12−14 lithium ion nucleator,6,15,16 interfacial layer,17,18 and modified separator19 were developed to regulate the Li deposition behavior. Lithiophilic materials such as graphene,20,21 graphene oxide,22 and porous carbon nanofiber23−27 have also been proposed to be host materials for composite Li anodes. Meanwhile, the Li anode has also been modified with the artificial solid electrolyte interphase (SEI) such as gel electrolyte28,29 and solid-state electrolytes30−33 to create a stable anodic interface. While significant progress has been © XXXX American Chemical Society

Received: November 16, 2018

A

DOI: 10.1021/acscentsci.8b00845 ACS Cent. Sci. XXXX, XXX, XXX−XXX

Research Article

ACS Central Science

Figure 1. Self-assembled monolayer passivated Li−C composite microparticles. (a) Schematic illustration of the molecular self-assembly process for the preparation of the OPA−Li−CNT sphere. (b, c) Scanning electron micrograph (SEM) of the OPA−Li−CNT spheres. (d) SEM and (e) phosphor elemental mapping of a focus-ion beam cut cross-section of an OPA−Li−CNT composite particle. XPS spectra of OPA passivated Li foil: (f) Li 1s spectrum, (g) P 2p spectrum, (h) C 1s spectrum, and (i) O 1s spectrum.

solvents. The resulting material, OPA−Li−CNT, not only exhibits high specific capacity (1850 mA h/g) and high CE (98.8%), but can also be processed in dry air and be slurrycoated. When paired with a commercial LiFePO4 cathode with an anode to cathode capacity ratio of 2:1, the slurry-coated Li anode retains 82.5% capacity after 250 cycles at 1C. Molecular Self-Assembly on Li. As illustrated in Figure 1a, the Li−CNT composite microparticles were prepared by impregnating spray-dried CNT microparticles with molten Li metal under mechanical stirring as previously reported.36,37 The Li−CNT powders were then stirred magnetically in an OPA solution for a few minutes. The resulting product, denoted as OPA−Li−CNT, displays very similar morphology and particle size distribution to that of Li−CNT microparticles (Figure 1b,c and Figure S1), while the energy-dispersive spectroscopy (EDS) mapping clearly shows the presence and the uniform distribution of phosphor element in the OPA−Li− CNT sphere (Figure S1e). The focus-ion-beam-milled (FIBmilled) cross-section of a particle and the corresponding EDS mapping reveal that the phosphor element mostly concentrates on the surface of the OPA−Li−CNT sphere (Figure 1d,e), which is indicative of the surface passivation nature of the process. X-ray photoelectron spectroscopy (XPS) was employed to characterize the chemical composition of the surface coating. The XPS spectra of OPA−Li−CNT are shown in Figure S2, and those of a piece of OPA passivated Li (OPA−Li) foil sample are shown in Figure 1. OPA−Li was used for this analysis to avoid interferences of the C signal from CNTs in OPA−Li−CNT. As shown in Figure 1f, the peaks at 54.5 and

slurry-coating while achieving excellent electrochemical performance. Here, we present the passivation of a Li−C composite material via self-assembly of octadecylphosphonic acid (OPA) molecules. The resulting material not only exhibits good electrochemical properties such as high specific capacity and high CE, but also becomes stable in dry air (dew point: −40 °C) and selected anhydrous solvents. These features allow the slurry-coating preparation of Li anodes with adjustable areal density.

■

RESULTS AND DISCUSSION The material is based on a micrometer-sized Li−carbon nanotube composite particle (Li−CNT) previously developed in our group.36,37 The Li−CNT composite microparticle exhibits high specific capacity and a good dendrite suppression property, but the small size also renders the material highly reactive. It can lead to violent spontaneous combustion when exposed to dry air, and therefore can only be processed strictly in a pure argon gas environment. For an improvement of the material processability as well as the CE in electrochemical cycling, Li−CNT particles are passivated with SAM as shown in Figure 1a. A passivation layer is formed on the surface of Li−CNT wherever lithium is exposed via self-assembly of OPA molecules. The structure of OPA consists of a saturated hydrocarbon chain and a phosphonic acid group that is reactive toward Li.32 The long alkyl chain provides the driving force for the assembly of a densely packed passivation layer on lithium metal, and its hydrophobic nature helps to block moisture in the environment or trace amounts of water in B

DOI: 10.1021/acscentsci.8b00845 ACS Cent. Sci. XXXX, XXX, XXX−XXX

Research Article

ACS Central Science

Figure 2. Dynamics study of the self-assembly process. Sum-frequency generation (SFG) vibrational spectra of the OPA assembly on a Li foil surface. (a) CH stretch region, and (b) OH stretch region. (c−e) Schematic illustration of the molecular self-assembly process.

times show a clear trend in both symmetric stretch and Fermi resonance modes in which the methylene signal appears first and then decreases, while the signal from the methyl group builds up as the methylene signal diminishes. Such a behavior is fully consistent with previously reported molecular self-assembly processes of alkyl reagents. During the early stage of the assembly (∼10 min, Figure 2d), OPA molecules are anchored to the Li surface via the lithium phosphate bonding, which raises the signals from CH2(ss) and CH2(FR) modes since the quantity of methylene chains at the Li surface increases. In the meantime, the surface density of OPA is low at this early stage, which allows a large degree of freedom for the octadecyl chain conformation, and therefore, the orientation of the terminal methyl group is randomized; thus, the signals from the methyl group are rather weak. As the SAM formation process proceeds, a higher density of OPA molecules on the Li surface leads to a densely packed SAM.50 The octadecyl chain becomes fully extended, and the neighboring methylene groups point to opposite directions; however, the terminal methyl groups are all aligned at the same tilt angle (Figure 2e). Therefore, the signals from methylene groups cancel out, but those from methyl groups build up. The fact that the ratio of CH3(FR) over CH2(FR) remains constant after 30 min indicates that the SAM formation is probably completed at 30 min. The major features observed in the O−H stretching region include the broad band at around 3300 cm−1, the relatively sharp peak at about 3010 cm−1, and the weak broad band at 3100 cm−1. It is well-accepted that the O−H stretching frequency reflects the local hydrogen bonding structure. The 3300 cm−1 band is similar to that observed for bulk water and most likely originates from the physically absorbed multilayer water,51 while the peak at 3010 cm−1, corresponding to OH groups with stronger hydrogen bonding, can be assigned to water species in close vicinity of the metal surface. Such OH groups may exist in a chemically reacted or strongly bound form, such as LiOH·H2O.52 Figure 2b reveals that there exist both strongly bound water and weakly adsorbed water on the

55.3 eV in the Li 1s spectrum are assigned to lithium and lithium phosphate in OPA−Li, respectively.38,39 The peaks at 133.5 and 131.5 eV in the P 2p spectrum are assigned to P in P−O and P−C bonds, respectively (Figure 1g).39,40 The peaks at ∼285.0 and 284.8 eV in the C 1s spectrum are usually assigned to C in C−P, C−C, or C−H bonds that are hardly distinguishable (Figure 1h).41,42 The major peak at 531.5 eV in the O 1s spectrum is assigned to O in phosphates (Figure 1i).43 The small sideband at 534 eV is similar to that in a previous study, which was interpreted as being from impurities or contaminations (Figure 1i).32 These results indicate that the coating layer on the Li surface is likely composed of lithium alkyl phosphate. It is worth noticing that the Li 1s and the P 2p spectra of the OPA−Li−CNT (Figure S2e,f) are essentially identical to those of OPA−Li foil in Figure 1f,g, clearly indicating the presence of the OPA layer. However, the C 1s and O 1s spectra are masked by high backgrounds from carbon nanotubes (Figure S2g,h). The dynamics of the self-assembly process is investigated with sum-frequency generation spectroscopy (SFG). SFG is an interfacial vibrational spectroscopy with a selection rule requiring that its signal arises from regions with broken centrosymmetry only, and thus has been widely used in studying molecular self-assembly processes such as alkyl thiols on gold,44,45 alkyl silane and alkyl phosphonic acid on silica, 46,47 and alkyl capping ligands on nanoparticle surfaces.48,49 Model samples of Li foil immersed in OPA solution for different times were rinsed and dried before measurement. The SFG vibrational spectra in the C−H and O−H stretching regions are shown in Figure 2a,b, respectively. The SFG spectra were recorded with the ssp polarization configuration. The characteristic vibrational modes at the C− H stretching region are methylene symmetric stretching (CH2(ss)) at 2855 cm−1, methyl symmetric stretching (CH3(ss)) at 2881 cm−1, and their corresponding methylene Fermi resonance modes (CH2(FR)) at 2910 cm−1 and methyl Fermi resonance (CH3(FR)) at 2940 cm−1, respectively.48,49 The series of Li foil samples treated with OPA for different C

DOI: 10.1021/acscentsci.8b00845 ACS Cent. Sci. XXXX, XXX, XXX−XXX

Research Article

ACS Central Science

Figure 3. Environmental stability of SAM passivated Li−C microparticles. Digital photographs of pouring different powder samples in a dry-air environment (dew point: −40 °C): (a) Li−CNT, (b) PA−Li−CNT, and (c) OPA−Li−CNT. XPS spectra of pristine and dry-air (dew point: −40 °C) exposed samples: (d) Li−CNT, (e) PA−Li−CNT, and (f) OPA−Li−CNT. (g) Electrochemical delithiation voltage profiles of pristine and dry-air exposed samples. (h) Electrochemical delithiation voltage profiles of the OPA−Li−CNT exposed to dry air for long hours. (i) Electrochemical delithiation voltage profiles of the OPA−Li−CNT after exposure to humid air (15 °C, 50% RH) for long hours.

displays no apparent reaction under the same condition (Video S3). Furthermore, the OPA−Li−CNT is also completely stable in air with 50% relative humidity (Video S4). These results indicate that the densely packed hydrophobic layer of octadecyl SAM is critical in sealing off moisture, oxygen, and nitrogen from Li metal. The stability of the OPA−Li−CNT in dry air is further confirmed with XPS, X-ray diffraction (XRD), and SEM. The Li 1s spectra in Figure 3d show that the Li element at the surface of the Li−CNT before exposure to dry air is essentially all Li metal (54.5 eV). After 1 h of exposure to dry air, a majority of surface Li metal is oxidized into Li2O (55.8 eV). The XRD pattern of the samples also indicates that the majority of the bulk Li metal has been oxidized into Li2O (Figure S3a). SEM shows that, after exposure to dry air for 1 h, the smooth morphology of the Li−CNT became irregular with grains (Figure S4a,d). The XPS spectra of both the OPA−Li− CNT and the control sample PA−Li−CNT display a major component of lithium phosphate (55.3 eV)39 and a minor component of Li metal before exposure. After 1 h of exposure to dry air the majority component at the surface of the PA− Li−CNT becomes Li2O, and the relative quantity of lithium phosphate is much reduced (Figure 3e). The XRD pattern of Pthe A−Li−CNT indicates that the majority of Li metal in the bulk is preserved, but the presence of Li2O is clearly made

pristine Li surface and Li surfaces with incomplete OPA SAM. However, both the 3010 and 3300 cm−1 bands disappear as a densely packed SAM is formed after 30 min, indicating that the surface hydroxide groups have reacted with OPA phosphonic groups, and a weakly adsorbed water layer has been expelled by the octadecyl chain as a result of the hydrophobic-interactiondriven self-assembly process. The weak broad band at around 3100 cm−1, which dominates upon the complete formation of OPA SAM, can be readily attributed to the OH groups brought by the OPA molecules. The expelling of water from the OPA SAM is crucial in improving the environmental stability and processability of the Li−C composite materials. Environmental Stability and Processability. The SAM passivated Li−C composite material OPA−Li−CNT exhibits excellent environmental stability and processability that are completely compatible with the current practice of battery manufacturing. It can be safely processed in dry air. Figure 3a− c shows photographs of three different powder samples being poured down in a dry-air environment with a −40 °C dew point. Untreated bare Li−CNT material reacts violently with dry air and spontaneously bursts into flames (Video S1). A control sample phosphoric acid passivated Li−CNT (PA−Li− CNT), which is Li−CNT treated with phosphoric acid, reacts with dry air less violently but still burnt red with heat into ashes (Video S2). In sharp contrast, the OPA−Li−CNT D

DOI: 10.1021/acscentsci.8b00845 ACS Cent. Sci. XXXX, XXX, XXX−XXX

Research Article

ACS Central Science

Figure 4. Slurry-coated lithium anode. (a) Digital photograph and (b) SEM micrograph of a slurry-coated lithium anode with areal capacity of 2.5 mA h/cm2 (the coating area shown is ∼8 cm in width and ∼15 cm length). Voltage profiles during Li stripping/plating cycles of Li||Li, Li−CNT|| Li−CNT and slurry-coated OPA−Li−CNT||OPA−Li−CNT cells in an ether-based electrolyte at current density of (c) 1 mA/cm2 and (d) 3 mA/ cm2. (e) Cycle performances of an OPA−Li−CNT||LFP coin cell. (f) Voltage profiles at different cycles of the OPA−Li−CNT∥LFP coin cell charged/discharged at 1C (1C = 170 mA h/g).

samples are all at 10 mV versus Li/Li+, the same as pure Li foils. The delithiation capacity of Li−CNT is 2070 mA h/g, and those of PA−Li−CNT and OPA−Li−CNT are 1843 and 1890 mA h/g, respectively. The slight reduction in specific capacity is probably due to surface reaction of Li with PA and OPA. After exposure to dry air (dew point: −40 °C) for 1 h, the specific capacity of Li−CNT, PA−Li−CNT, and OPA− Li−CNT becomes 885, 1317, and 1790 mA h/g, corresponding to specific capacity retention of 42.7%, 71.4%, and 94.7%, respectively. The OPA−Li−CNT sample is particularly stable. It displays a delithiation capacity of 1734 and 1705 mA h/g after 72 h and 1 week, respectively, in dry air (Figure 3h). OPA−Li−CNT can even withstand humid air (15 °C, 50% RH): the delithiation capacity is observed to be 1685 mA h/g after exposure to humid air for 2 h and 1050 mA h/g after 72 h (Figure 3i). Such high Li storage capacity, high stability, and excellent processability provide a solid foundation for exploring the electrochemical performance of the OPA−Li−CNT toward practical applications. Slurry-Coated Lithium Anode. In spite of the small particle size (∼5 μm), the OPA−Li−CNT is stable in dry air and selected anhydrous solvents thanks to SAM passivation,

evident (Figure S3b). The SEM result shows that a morphological change of the PA−Li−CNT after exposure to dry air is much less severe compared to the Li−CNT (Figure S4b,e). By contrast, the XPS spectra indicate that the surface of the OPA−Li−CNT is largely unchanged with only very little amount of Li metal converted to Li2O (Figure 3f), and there is virtually no Li2O detected in the bulk of the OPA−Li−CNT via XRD after 1 h of exposure in dry air (Figure S3c); moreover, there is almost no morphological change after 1 h of exposure in dry air (Figure S4c,f). The effect of OPA on the environmental stability of Li metal foil was also studied for comparison. The Li 1s spectra in Figure S5a confirmed that lithium metal at the surface of lithium foil has been partially oxidized into Li2O after 60 min of exposure in dry air (dew point: −40 °C). In contrast, the OPA passivated Li foil maintains its surface chemical composition with only a trace amount of Li2O detected after the same exposure to the dry-air environment (Figure S5b). The electrochemical delithiation test is used to quantify the material stability. Various samples are pressed onto copper foam and directly used as electrodes for delithiation experiments. As shown in Figure 3g, the delithiation potentials of the E

DOI: 10.1021/acscentsci.8b00845 ACS Cent. Sci. XXXX, XXX, XXX−XXX

Research Article

ACS Central Science

SEI. Furthermore, the stability of the OPA SAM layer in organic electrolyte was also investigated by measuring the evolution of impedance spectra over storage time. As shown in Figure S10, the impedance of Li−CNT increased dramatically from ∼20 to ∼100 Ω over 120 h of storage, while the impedance of the OPA−Li−CNT symmetric cell increased slowly from ∼30 to ∼50 Ω over the same storage time. Considering the local curvature of the OPA−Li−CNT particles, the increase in impedance could be attributed to the imperfect coverage of lithium by OPA. Thus, the OPA SAM layer is stable and does not dissolve in organic electrolyte. The protective effect of the OPA SAM layer during cycling is further investigated via comparison of postmortem SEM images against those of lithium metal foil and Li−CNT electrodes. Figure S11 shows that, after 200 cycles, coarse filament-like dendrites several μm in size were observed on the Li foil electrode. Li−CNT particles maintained their spherical shape after 200 cycles without any dendrites, but the particles appeared to be covered with a new layer. In contrast, the morphologies of OPA−Li−CNT particles were well-maintained after 200 cycles without apparent dendrites or extra coating, indicating that the OPA layer had protected the decomposition of electrolyte on the surface of the Li−CNT to a good extent. These electrochemical behaviors of the OPA− Li−CNT material suggest that the OPA SAM layer not only leads to excellent environmental stability and processability but may also improve the CE toward practical applications. Upon pairing with a LiFePO4 cathode (1.25 mA h/cm2, the diameter of the electrode is 15 mm) to form a full cell at a cathode to anode capacity ratio of 1:2 and cycled at 1C between 2.5 and 4.1 V in an ether-based electrolyte, more than 250 stable cycles were recorded before rapid capacity decay started (Figure 4e). Taking this point as the complete consumption of the lithium from the anode, a CE of ∼98.8% is calculated (for details of the calculation, please refer to the SI). This CE of the slurrycoated Li anode is ∼4.8% higher than that of the Li−CNT anode prepared by pressing the powders on a copper foam and ∼6.3% higher than that of the conventional Li foil anode measured at the same condition.36,37 Figure 4f shows the voltage profiles of the OPA−Li−CNT||LFP cell at various cycles. Even though the capacity of the cell gradually decreases with cycling, the overpotential of the voltage profiles keeps almost constant at ∼80 mV during cycling, which indicates good stability of the OPA−Li−CNT material in the LFP full battery. Li metal materials, if successfully utilized in batteries, can be deployed as the lithiation reagent in lithium ion batteries to compensate for the irreversible loss of capacity in the initial cycle, or they can be paired with lithium-free cathodes in “beyond lithium ion” battery chemistries such as Li−S and Li− O2. Ideal Li metal materials for practical utilization not only shall display excellent electrochemical performances required by their respective applications, but also need to fulfill the following additional requirements: (1)The production, including the fabrication of the host material and the subsequent lithiation of the host, needs to be scalable to industrial quantities and also cost-effective. (2) While the production of Li metal materials may take place in Ar environment, they need to be safe and processable in a dry-air environment, which is a standard in the battery manufacturing industry. (3) The utilization of the material should be compatible with the current battery manufacturing processes.

which enables the slurry-coating preparation of Li anodes with controllable areal mass loading. As shown in Figure 4a, an OPA−Li−CNT anode has been slurry-coated on a Cu foil current collector with areal capacity of 2.5 mA h/cm2 (the thickness is measured to be ∼200 μm, Figure S7), using SBR/ PS (10 wt %) as binder, AB (10 wt %) as electronic conducting additive, and p-xylene as solvent. SEM images in Figure 4b show that the OPA−Li−CNT particles maintain their spherical shap,e and the XPS spectra in Figure S6b showed no chemical composition changes on the surface of the OPA− Li−CNT after slurry-coating. The electrochemical performance of the slurry-coated OPA−Li−CNT anode was tested in a symmetric OPA−Li− CNT||OPA−Li−CNT cell cycled in an ether-based electrolyte consisting of 1 M LiTFSI in a solvent mixture of DOL/DME (DOL, 1,3-dioxolane; 1:1 vol %). Symmetric cells of Li−CNT|| Li−CNT using Cu foam-based Li−CNT electrodes and Li||Li using lithium foil as electrodes are assembled and measured for comparison. The Li−CNT electrode was made using Cu foam as current collector but not slurry-coating since it reacts with the solvent and yields Li2CO3 (Figure S6a). Although a self-assembled alkyl layer may be expected to hinder the Li stripping/platting, the OPA−Li−CNT||OPA− Li−CNT cell exhibited a smaller and more stable overpotential. When cycled at a current density of 1 mA/cm2 (Figure 4c), the OPA−Li−CNT||OPA−Li−CNT cell exhibited an overpotential of ∼25 mV (Figure S8a) at the beginning of the cycling, and the value slowly increase to ∼45 mV in 200 cycles (Figure S8c). In sharp contrast, the Li− CNT||Li−CNT cell showed an overpotential of ∼60 mV at the beginning of the cycling (Figure S8a), and then, the value rose quickly to ∼95 mV within the first 200 cycles (Figure S8c). The Li||Li cell demonstrated a typical stripping/plating behavior of lithium metal, which showed an overpotential of ∼75 mV at the beginning of the cycling (Figure S8a) which then drops to 70 mV after 100 cycles (Figure S8b) because of the increase of surface area contributed by lithium dendrite; then, the value rose quickly, and the overpotential reached ∼175 mV at the end of 200 cycles (Figure S8c). At a higher current density of 3 mA/cm2 (Figure 4d), greater overpotential values were observed for all three cells. However, the overpotentials of the OPA−Li−CNT||OPA−Li−CNT and Li−CNT||Li−CNT cell were much more stable compared to that of the Li||Li cell during 200 cycles. It is worth noting that the depth of discharge (DOD) of Li stripping/platting is 20% for the Li−CNT||Li−CNT and OPA−Li−CNT||OPA−Li− CNT cell, while the DOD of the Li||Li cell (thickness of the lithium foil: 100 μm) is only 3% since it is difficult to prepare a very thin lithium foil. The electrochemical stability of the OPA−Li−CNT was confirmed with electrochemical impedance spectroscopy (EIS) measurements. As shown in the Nyquist plots in Figure S9, the impedance of the Li foil||Li foil cell first decreased from 40 to 27 Ω after 50 cycles, which could be attributed to the increase of specific surface area of the lithium electrode due to dendrite growth. Then, the impedance increased drastically to ∼110 Ω after 200 cycles, which could be attributed to the aggregation of “dead” lithium. The impedance of the Li−CNT||Li−CNT cell slowly increased from ∼25 to ∼50 Ω at the 200th cycle. In sharp contrast, the impedance of the OPA−Li−CNT||OPA− Li−CNT cell remained relatively stable and increased by only ∼8 Ω in 200 cycles, indicating that the OPA layer can prevent the electrolyte from reacting with Li to form a high-impedance F

DOI: 10.1021/acscentsci.8b00845 ACS Cent. Sci. XXXX, XXX, XXX−XXX

Research Article

ACS Central Science

Preparation of PA−Li−CNT Composite. In a typical synthesis of the OPA−Li−CNT composite, 200 mg of asprepared Li−CNT was dispersed in 20 mL of a 0.1 wt % PA in tetrahydrofuran (THF) solution by magnetic stirring for several minutes. The resultant solution was filtered after rinsing with THF several times and dried at 333 K for 10 h under vacuum. Materials Characterization. The sample morphology was characterized using an SEM instrument (Hitachi S-4800) operated at 3 kV. Li−CNT and Li foil electrodes were transferred from a glovebox to an SEM chamber via a homebuilt gastight transfer device to protect the sample from ambient air. The chemical elements present in the sample were analyzed by EDS (Quanta FEG 250) operated at 20 kV. XPS spectra were taken with a Thermo Scientific ESCA Lab 250Xi instrument using 200 W monochromated Al Kα radiation. The samples were transferred from an Ar-filled glovebox to the XPS measurement chamber in a custom-made XPS transfer chamber with gastight sealing (Figure S12). The FIB milling and the corresponding SEM imaging of the cross-section samples were conducted in a dual-beam Nova 200 NanoLab UHRFEG system. The SFG spectroscopy was carried out on an EKSPLA system with a copropagating configuration. The XRD pattern of various samples was obtained on a Bruker D8 diffractometer using Cu Kα radiation. Electrode Fabrication. For the powder press method, Li− CNT and OPA−Li−CNT powders were pressed on a roundshaped copper foam (1.75 cm in diameter and 600 μm in thickness with the pore size of 60−70 μm) and directly used as electrodes. The areal capacity is ∼20 mA h/cm2. For the slurry-coating method, 20 mg of poly(styrene-cobutadiene) (SBR) (Sigma-Aldrich) and 20 mg of polystyrene (PS) (molecular weight: 2 000 000, Alfa Aesar) were dissolved in 1500 μL of p-xylene (anhydrous, 99.5%) (Sigma-Aldrich) followed by addition of 40 mg of acetylene black, and 320 mg of OPA−Li−CNT. After magnetic stirring for 10 h, the slurry was coated with a doctor blade on a copper foil with the mass loading (active material) of 2.5 mg/cm2, and then, the electrodes were dried in a vacuum oven at 333 K for 10 h. Electrochemical Characterization. The 2025 type coin cells were used to fabricate batteries for all electrochemical measurements. Polypropylene membrane (Celgard 2400) was used as the separator. The EIS measurements were performed on an RST 5000 electrochemical workstation (Suzhou Risetest Electronic Co., Ltd.) over the frequency range from 100 mHz to 100 kHz with an amplitude of 5 mV. Electrochemical performance was evaluated by a NEWARE battery analyzer (NEWARE technology Ltd., Shenzhen, China). Stripping/ plating experiments were performed on symmetric cells of Li foil, Li−CNT, and OPA−Li−CNT electrodes at 1 and 3 mA/ cm2 for 200 cycles, with each cycle involving a capacity of 0.5 mA h/cm2. The electrolyte consisted of 1 M LiTFSI in a solvent mixture of DOL/DME (1:1 vol %). In the full cell testing, OPA−Li−CNT electrode with a capacity of 2.5 mA h/ cm2 and LFP (Sinlion Battery Tech, Co., Ltd.) electrode with a capacity of 1.25 mA h/cm2 were used as anode and cathode, respectively. The electrolyte consisted of 1 M LiTFSI in a solvent mixture of DOL/DME (1:1 vol %) with 2 wt % LiNO3 as additive. Electrochemical Delithiation. For an evaluation of the specific capacity of the OPA−Li−CNT composite, 25 mg of the OPA−Li−CNT composite was assembled in a Li||OPA− Li−CNT battery. The battery was galvanostatically charged at

Our results show that a long-chain aliphatic SAM passivation layer can effectively protect reactive Li metal materials from water, oxygen, and electrolyte, and at the same time conducting Li ion. Consequently, the OPA−Li−CNT material can be safely handled in dry or even humid air and could be slurry-coated with organic solvents. Since the raw materials used for preparing the OPA−Li−CNT material (i.e., CNT and OPA) are available in large quantities and low cost, and the manufacturing techniques, i.e., spray drying of CNT dispersion, the molten impregnation for prelithiation of the CNT microspheres, and the OPA passivation of Li−CNT, are all scalable, this material has a good chance for potential applications. Further research on the improvement of CE via stabilization of the SAM in electrochemical cycling is currently underway.

■

CONCLUSION In conclusion, we have demonstrated the molecular selfassembly of OPA passivated layer on a highly reactive Li− CNT composite microsphere to address its safety, environmental stability, and processability issues. SFG was used to investigate the dynamics of the self-assembly process. The resulting OPA−Li−CNT material displays excellent environmental stability while significantly improving the electrochemical performance. It is stable in dry air and can even be exposed to 50% humid air without catching fire. Such robustness allows the material to be processed into anodes using the conventional slurry-coating procedure with organic solvents. When the slurry-coated OPA−Li−CNT electrode is paired with an LFP cathode at a 1:2 cathode to anode capacity ratio, over 250 stable cycles were observed, which corresponds to a CE of 98.8%. The study of molecular self-assembly for Li metal passivation may also provide inspiration to the development of other alkali metal materials for battery applications.

■

METHODS

Preparation of CNT Particles. In a typical synthesis of CNT particles, 10.0 g of CNT (Shandong Dazhan Nano Materials Technology Co., Ltd. GT-400) was dispersed in a mixture of 1000 mL of ethanol and 100 mL of deionized water, and then ultrasonicated for 2 h. The obtained CNT suspension was then dried using a spray dryer. The inlet and outlet temperatures of spray dryer were fixed at 473 and 373 K, respectively, and the flow rate of air was maintained at 5 L/min at a feeding speed of 500 mL/h. Subsequently, the CNT particles were collected from collector of the spray dryer. Preparation of Li−CNT Composite. The Li−CNT composite was prepared in an Ar-filled glovebox by mixing 4 g of Li foil and 2 g of CNT particles in a 200 mL reactor made of stainless steel. The temperature of the reactor was raised to 493 K. Then, the molten Li and CNT particles were mechanically mixed for 20 min, and naturally cooled to room temperature. The obtained Li−CNT composite was further used to prepare the OPA−Li−CNT composite. Preparation of OPA−Li−CNT Composite. In a typical synthesis of the OPA−Li−CNT composite, 200 mg of asprepared Li−CNT was dispersed in 20 mL of a 0.1 wt % OPA in tetrahydrofuran (THF) solution by magnetic stirring for several minutes. The resultant solution was filtered after rinsing with THF several times and dried at 333 K for 10 h under vacuum. G

DOI: 10.1021/acscentsci.8b00845 ACS Cent. Sci. XXXX, XXX, XXX−XXX

ACS Central Science

■

a current density of 0.5 mA/cm2 until all of the lithium was electrochemically stripped from the OPA−Li−CNT composite.

■

REFERENCES

(1) Guo, Y.; Li, H.; Zhai, T. Reviving lithium-metal anodes for nextgeneration high-energy batteries. Adv. Mater. 2017, 29, 1700007. (2) Lin, D.; Liu, Y.; Cui, Y. Reviving the lithium metal anode for high-energy batteries. Nat. Nanotechnol. 2017, 12, 194−206. (3) Yang, C.; Fu, K.; Zhang, Y.; Hitz, E.; Hu, L. Protected lithiummetal anodes in batteries: from liquid to solid. Adv. Mater. 2017, 29, 1701169. (4) Cheng, X. B.; Zhang, R.; Zhao, C. Z.; Wei, F.; Zhang, J. G.; Zhang, Q. A review of solid electrolyte interphases on lithium metal anode. Adv. Sci. (Weinh) 2016, 3, 1500213. (5) Zhang, X.-Q.; Cheng, X.-B.; Chen, X.; Yan, C.; Zhang, Q. Fluoroethylene carbonate additives to render uniform li deposits in lithium metal batteries. Adv. Funct. Mater. 2017, 27, 1605989. (6) Zhang, R.; Chen, X. R.; Chen, X.; Cheng, X. B.; Zhang, X. Q.; Yan, C.; Zhang, Q. Lithiophilic sites in doped graphene guide uniform lithium nucleation for dendrite-free lithium metal Anodes. Angew. Chem., Int. Ed. 2017, 56, 7764−7768. (7) Cheng, X. B.; Zhang, R.; Zhao, C. Z.; Zhang, Q. Toward safe lithium metal anode in rechargeable batteries: a review. Chem. Rev. 2017, 117, 10403−10473. (8) Lu, Y.; Das, S. K.; Moganty, S. S.; Archer, L. A. Ionic liquidnanoparticle hybrid electrolytes and their application in secondary lithium-metal batteries. Adv. Mater. 2012, 24, 4430−4435. (9) Ding, F.; Xu, W.; Graff, G. L.; Zhang, J.; Sushko, M. L.; Chen, X.; Shao, Y.; Engelhard, M. H.; Nie, Z.; Xiao, J.; Liu, X.; Sushko, P. V.; Liu, J.; Zhang, J. G. Dendrite-free lithium deposition via self-healing electrostatic shield mechanism. J. Am. Chem. Soc. 2013, 135, 4450− 4456. (10) Zhang, Y.; Qian, J.; Xu, W.; Russell, S. M.; Chen, X.; Nasybulin, E.; Bhattacharya, P.; Engelhard, M. H.; Mei, D.; Cao, R.; Ding, F.; Cresce, A. V.; Xu, K.; Zhang, J. G. Dendrite-free lithium deposition with self-aligned nanorod structure. Nano Lett. 2014, 14, 6889−6896. (11) Zheng, J.; Engelhard, M. H.; Mei, D.; Jiao, S.; Polzin, B. J.; Zhang, J.-G.; Xu, W. Electrolyte additive enabled fast charging and stable cycling lithium metal batteries. Nature Energy 2017, 2, 17012. (12) Liu, W.; Lin, D.; Pei, A.; Cui, Y. Stabilizing lithium metal anodes by uniform Li-ion flux distribution in nanochannel confinement. J. Am. Chem. Soc. 2016, 138, 15443−15450. (13) Yang, C. P.; Yin, Y. X.; Zhang, S. F.; Li, N. W.; Guo, Y. G. Accommodating lithium into 3D current collectors with a submicron skeleton towards long-life lithium metal anodes. Nat. Commun. 2015, 6, 8058. (14) Ye, H.; Xin, S.; Yin, Y. X.; Li, J. Y.; Guo, Y. G.; Wan, L. J. Stable Li plating/stripping electrochemistry realized by a hybrid Li reservoir in spherical carbon granules with 3D conducting skeletons. J. Am. Chem. Soc. 2017, 139, 5916−5922. (15) Yang, C.; Yao, Y.; He, S.; Xie, H.; Hitz, E.; Hu, L. Ultrafine silver nanoparticles for seeded lithium deposition toward stable lithium metal anode. Adv. Mater. 2017, 29, 1702714. (16) Yan, K.; Lu, Z.; Lee, H.-W.; Xiong, F.; Hsu, P.-C.; Li, Y.; Zhao, J.; Chu, S.; Cui, Y. Selective deposition and stable encapsulation of lithium through heterogeneous seeded growth. Nature Energy 2016, 1, 16010. (17) Liang, Z.; Zheng, G.; Liu, C.; Liu, N.; Li, W.; Yan, K.; Yao, H.; Hsu, P.-C.; Chu, S.; Cui, Y. Polymer nanofiber-guided uniform lithium deposition for battery electrodes. Nano Lett. 2015, 15, 2910− 2916. (18) Zheng, G.; Lee, S. W.; Liang, Z.; Lee, H. W.; Yan, K.; Yao, H.; Wang, H.; Li, W.; Chu, S.; Cui, Y. Interconnected hollow carbon nanospheres for stable lithium metal anodes. Nat. Nanotechnol. 2014, 9, 618−623. (19) Liu, Y.; Liu, Q.; Xin, L.; Liu, Y.; Yang, F.; Stach, E. A.; Xie, J. Making Li-metal electrodes rechargeable by controlling the dendrite growth direction. Nature Energy 2017, 2, 17083. (20) Zhang, R.; Cheng, X. B.; Zhao, C. Z.; Peng, H. J.; Shi, J. L.; Huang, J. Q.; Wang, J.; Wei, F.; Zhang, Q. Conductive nanostructured scaffolds render low local current density to inhibit lithium dendrite growth. Adv. Mater. 2016, 28, 2155−2162.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acscentsci.8b00845.

■

Research Article

Calculation of Coulombic efficiency and additional figures including SEM images, diameter distributions, XPS spectra, XRD results, overpotentials, Nyquist plots, photograph, and schematics (PDF) Video S1: untreated bare Li−CNT material reacts violently with dry air and spontaneously bursts into flames (AVI) Video S2: PA−Li−CNT reacts with dry air less violently but still burns red with heat into ashes (AVI) Video S3: OPA−Li−CNT displays no apparent reaction under the same conditions (AVI) Video S4: OPA−Li−CNT is also completely stable in air with 50% relative humidity (AVI)

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. *E-mail: [email protected]. ORCID

Tuo Kang: 0000-0002-1868-3140 Yejun Qiu: 0000-0003-0519-0424 Yanbin Shen: 0000-0002-1278-1137 Liwei Chen: 0000-0003-4160-9771 Author Contributions

L.C. conceived the original idea. T.K., Y.Q., Y.S., and L.C. designed all the experiments. T.K., Y.W., F.G., and W.L. prepared the samples. T.K. and J.Z. slurry-coated the OPA− Li−CNT electrodes. T.K. and C.L. performed the stability and electrochemical tests. T.K., J.Y., and H.L. carried out the SFG characterization. T.K., Y.Q., Y.S., and L.C. wrote the paper. All authors were involved in analysis of the experimental data, discussion of the results, and the preparation of the manuscript. Notes

The authors declare no competing financial interest. Safety statement: Li−CNT particles spontaneously react with dry or moist air, and thus must be handled with care in Ar atmosphere. OPA−Li−CNT can be processed in dry and moist air. Waste Li−CNT and OPA−Li−CNT materials need to be slowly oxidized under controlled conditions before disposed as chemical wastes.

■

ACKNOWLEDGMENTS This work was financially supported by the Ministry of Science and Technology (Grant 2016YFB0100102), the “Strategic Priority Research Program” of the CAS (Grants XDA09010600 and XDA09010303), and the National Natural Science Foundation of China (Grants 21625304, 21733012, and 21773290). H

DOI: 10.1021/acscentsci.8b00845 ACS Cent. Sci. XXXX, XXX, XXX−XXX

Research Article

ACS Central Science (21) Liu, S.; Wang, A.; Li, Q.; Wu, J.; Chiou, K.; Huang, J.; Luo, J. Crumpled graphene balls stabilized dendrite-free lithium metal anodes. Joule 2018, 2, 184−193. (22) Lin, D.; Liu, Y.; Liang, Z.; Lee, H. W.; Sun, J.; Wang, H.; Yan, K.; Xie, J.; Cui, Y. Layered reduced graphene oxide with nanoscale interlayer gaps as a stable host for lithium metal anodes. Nat. Nanotechnol. 2016, 11, 626−632. (23) Zhang, Y.; Wang, C.; Pastel, G.; Kuang, Y.; Xie, H.; Li, Y.; Liu, B.; Luo, W.; Chen, C.; Hu, L. 3D Wettable framework for dendritefree alkali metal anodes. Adv. Energy Mater. 2018, 8, 1800635. (24) Liu, L.; Yin, Y.-X.; Li, J.-Y.; Li, N.-W.; Zeng, X.-X.; Ye, H.; Guo, Y.-G.; Wan, L.-J. Free-standing hollow carbon fibers as high-capacity containers for stable lithium metal anodes. Joule 2017, 1, 563−575. (25) Zhang, R.; Chen, X.; Shen, X.; Zhang, X.-Q.; Chen, X.-R.; Cheng, X.-B.; Yan, C.; Zhao, C.-Z.; Zhang, Q. Coralloid carbon fiberbased composite lithium anode for robust lithium metal batteries. Joule 2018, 2, 764−777. (26) Liu, Y.; Lin, D.; Liang, Z.; Zhao, J.; Yan, K.; Cui, Y. Lithiumcoated polymeric matrix as a minimum volume-change and dendritefree lithium metal anode. Nat. Commun. 2016, 7, 10992. (27) Lang, J.; Jin, Y.; Luo, X.; Liu, Z.; Song, J.; Long, Y.; Qi, L.; Fang, M.; Li, Z.; Wu, H. Surface graphited carbon scaffold enables simple and scalable fabrication of 3D composite lithium metal anode. J. Mater. Chem. A 2017, 5, 19168−19. (28) Zheng, G.; Wang, C.; Pei, A.; Lopez, J.; Shi, F.; Chen, Z.; Sendek, A. D.; Lee, H.-W.; Lu, Z.; Schneider, H.; Safont-Sempere, M. M.; Chu, S.; Bao, Z.; Cui, Y. High-performance lithium metal negative electrode with a soft and flowable polymer coating. ACS Energy Letters 2016, 1, 1247−1255. (29) Wu, H.; Cao, Y.; Geng, L.; Wang, C. In situ formation of stable interfacial coating for high performance lithium metal anodes. Chem. Mater. 2017, 29, 3572−3579. (30) Xie, J.; Sendek, A. D.; Cubuk, E. D.; Zhang, X.; Lu, Z.; Gong, Y.; Wu, T.; Shi, F.; Liu, W.; Reed, E. J.; Cui, Y. Atomic layer deposition of stable LiAlF4 lithium ion conductive interfacial layer for stable cathode cycling. ACS Nano 2017, 11, 7019−7027. (31) Lin, D.; Liu, Y.; Chen, W.; Zhou, G.; Liu, K.; Dunn, B.; Cui, Y. Conformal lithium fluoride protection layer on three-dimensional lithium by nonhazardous gaseous reagent freon. Nano Lett. 2017, 17, 3731−3737. (32) Li, N. W.; Yin, Y. X.; Yang, C. P.; Guo, Y. G. An artificial solid electrolyte interphase layer for stable lithium metal anodes. Adv. Mater. 2016, 28, 1853−1858. (33) Li, X.; Guo, S.; Deng, H.; Jiang, K.; Qiao, Y.; Ishida, M.; Zhou, H. An ultrafast rechargeable lithium metal battery. J. Mater. Chem. A 2018, 6, 15517−15522. (34) Ai, G.; Wang, Z.; Zhao, H.; Mao, W.; Fu, Y.; Yi, R.; Gao, Y.; Battaglia, V.; Wang, D.; Lopatin, S.; Liu, G. Scalable process for application of stabilized lithium metal powder in Li-ion batteries. J. Power Sources 2016, 309, 33−41. (35) Ryou, M.-H.; Lee, Y. M.; Lee, Y.; Winter, M.; Bieker, P. Mechanical surface modification of lithium metal: towards improved Li metal anode performance by directed Li plating. Adv. Funct. Mater. 2015, 25, 834−841. (36) Guo, F.; Wang, Y.; Kang, T.; Liu, C.; Shen, Y.; Lu, W.; Wu, X.; Chen, L. A Li-dual carbon composite as stable anode material for Li batteries. Energy Storage Materials 2018, 15, 116−123. (37) Wang, Y.; Shen, Y.; Du, Z.; Zhang, X.; Wang, K.; Zhang, H.; Kang, T.; Guo, F.; Liu, C.; Wu, X.; Lu, W.; Chen, L. A lithium-carbon nanotube composite for stable lithium anodes. J. Mater. Chem. A 2017, 5, 23434−23439. (38) Contour, J. P.; Salesse, A.; Froment, M.; Garreau, M.; Thevenin, J.; Warin, D. Analysis by electron-microscopy and xps of lithium surfaces polarized in anhydrous organic electrolytes. Journal De Microscopie Et De Spectroscopie Electroniques 1979, 4, 483−491. (39) Morgan, W. E.; Stec, W. J.; Vanwazer, J. R. Inner-orbital photoelectron spectroscopy of alkali-metal halides, perchlorates, phosphates, and pyrophosphates. J. Am. Chem. Soc. 1973, 95, 751− 755.

(40) Shul'ga, Y. M.; Bulatov, A. V.; Gould, R. A. T.; Konze, W. V.; Pignolet, L. H. X-ray photoelectron-spectroscopy of a series of heterometallic gold platinum phosphine cluster compounds. Inorg. Chem. 1992, 31, 4704−4706. (41) Swiatowska, J.; Lair, V.; Pereira-Nabais, C.; Cote, G.; Marcus, P.; Chagnes, A. XPS, XRD and SEM characterization of a thin ceria layer deposited onto graphite electrode for application in lithium-ion batteries. Appl. Surf. Sci. 2011, 257, 9110−9119. (42) Larsson, R.; Malek, A.; Folkesson, B. Ir intensity measurements on phenyl groups for esca calibration and the estimation of effective charges applied to phosphorus-compounds. Chemica Scripta 1980, 16, 47−51. (43) Dang, T. A.; Chau, C. N. Electron spectroscopy for chemical analysis of cool white phosphors coated with SiO2 thin film. J. Electrochem. Soc. 1996, 143, 302−305. (44) Yeganeh, M. S.; Dougal, S. M.; Polizzotti, R. S.; Rabinowitz, P. Interfacial atomic structure of a self-assembled alkyl thiol monolayer/ Au(111): A sum-frequency generation study. Phys. Rev. Lett. 1995, 74, 1811−1814. (45) Jacob, J. D. C.; Lee, T. R.; Baldelli, S. In situ vibrational study of the reductive desorption of alkanethiol monolayers on gold by sum frequency generation spectroscopy. J. Phys. Chem. C 2014, 118, 29126−29134. (46) Dubey, M.; Weidner, T.; Gamble, L. J.; Castner, D. G. Structure and order of phosphonic acid-based self-assembled monolayers on Si(100). Langmuir 2010, 26, 14747−14754. (47) Liu, Y.; Wolf, L. K.; Messmer, M. C. A study of alkyl chain conformational changes in self-assembled n-octadecyltrichlorosilane monolayers on fused silica surfaces. Langmuir 2001, 17, 4329−4335. (48) Roke, S.; Berg, O.; Buitenhuis, J.; van Blaaderen, A.; Bonn, M. Surface molecular view of colloidal gelation. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 13310−13314. (49) Zhang, H.; Li, F.; Xiao, Q.; Lin, H. Conformation of capping ligands on nanoplates: facet-edge-induced disorder and self-assemblyrelated ordering revealed by sum frequency generation Spectroscopy. J. Phys. Chem. Lett. 2015, 6, 2170−2176. (50) Weeraman, C.; Yatawara, A. K.; Bordenyuk, A. N.; Benderskii, A. V. Effect of nanoscale geometry on molecular conformation: vibrational sum-frequency generation of alkanethiols on gold nanoparticles. J. Am. Chem. Soc. 2006, 128, 14244−14245. (51) Daldrop, J. O.; Saita, M.; Heyden, M.; Lorenz-Fonfria, V. A.; Heberle, J.; Netz, R. R. Orientation of non-spherical protonated water clusters revealed by infrared absorption dichroism. Nat. Commun. 2018, 9, 311. (52) Zhang, X.; Guo, L.; Gan, L.; Zhang, Y.; Wang, J.; Johnson, L. R.; Bruce, P. G.; Peng, Z. LiO2: cryosynthesis and chemical/ electrochemical reactivities. J. Phys. Chem. Lett. 2017, 8, 2334−2338.

I

DOI: 10.1021/acscentsci.8b00845 ACS Cent. Sci. XXXX, XXX, XXX−XXX