In Situ XRD and TEM Studies of Sol-Gel-Based ... - ACS Publications

Subscriber access provided by United Arab Emirates University | Libraries Deanship

Article 4

In Situ XRD and TEM Studies of Sol-Gel-based Synthesis of LiFePO Dominika Agnieszka Ziolkowska, Jacek B. Jasinski, Bartosz Hamankiewicz, Krzysztof P. Korona, She-Huang Wu, and Andrzej Czerwinski

Cryst. Growth Des., Just Accepted Manuscript • DOI: 10.1021/acs.cgd.6b00575 • Publication Date (Web): 09 Aug 2016 Downloaded from http://pubs.acs.org on August 16, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Crystal Growth & Design is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

In Situ XRD and TEM Studies of Sol-Gel-based Synthesis of LiFePO4 Dominika A. Ziolkowskaa,*, Jacek B. Jasinskib,*, Bartosz Hamankiewiczc, Krzysztof P. Koronaa, She-Huang Wud and Andrzej Czerwinskic a

b

Faculty of Physics, University of Warsaw, Pasteura 5, 02-093 Warsaw, Poland Conn Center for Renewable Energy Research, University of Louisville, Louisville, KY

40292, USA c

Faculty of Chemistry, University of Warsaw, Pasteura 1, 02-093 Warsaw, Poland

d

Department of Materials Engineering, Tatung University, No.40, 3rd Sec, Zhongshan N.

Rd., Taipei 104, Taiwan

KEYWORDS: in situ TEM, in situ XRD, lithium ion batteries, sol-gel method, cathodes

ABSTRACT: Parallel in situ TEM and XRD heating experiments of LiFePO4 precursors obtained by sol-gel method were conducted to study changes and to understand structural and morphological evolution during synthesis annealing, which is one of the most critical

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 33

stages in preparing rechargeable cathodes based on these materials. Raman spectroscopy and electrochemical testing were also performed and a basic optimization of the final step of the sol-gel process was demonstrated by comparing in situ heating data with the electrochemical performance of materials annealed at different temperatures. The results obtained from these in situ measurements, at different length scales, provided a detailed picture of the structural and morphological changes and provided a better understanding of the electrochemical behavior of the final LiFePO4 material. The study showed a strong dependence between the electrochemical performance of LiFePO4 synthesized by sol-gel method and annealing temperature. The best performance was obtained with a material annealed at 800°C.

Introduction In recent years, the lithium-ion battery (LIB) technology has been widely used for various energy storage applications, from portable consumer electric and electronic devices, to hybrid and electric cars, to power grid applications and efficient integration of renewable energy sources.1,2 Among various materials proposed for LIB cathodes, lithium phospho-olivine compounds are some of the most promising because of their low cost, good thermal stability, high theoretical capacity of 170 mAh g-1, and high open circuit voltage of 3.5 V. These characteristics make lithium phospho-olivine compounds viable candidates to replace presently used cobalt oxide electrodes.3-5 So far, there have been only a few in situ electrochemical studies of lithiation/delithiation processes in lithium iron phosphate (LiFePO4) reported, most of


2

Page 3 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


them based on synchrotron X-ray sources6-14 or, in the case of recent works15-17, focused on in situ transmission electron microscopy (TEM) electrochemical measurements. Similarly, there have been only a few reports on in situ annealing of LiFePO4, conducted using TEM18-21 or X-ray diffraction (XRD)22-24 on LiFePO4 precursors, prepared by wet chemistry or ball-milling methods. The importance of in situ studies is that they can provide direct information on the structural and morphological evolution of the material during its synthesis, which in turn illustrates optimal synthesis conditions and improves the efficiency of the process. In situ real-time experiments enable direct observation of the sample under dynamic conditions showing effects which normally are not discernible during ex situ post-synthesis measurements. In particular, in situ TEM studies, where simultaneous imaging, spectroscopic, and diffraction measurements can be conducted with high spatial resolution, often at the nanoscale, are ideal tools for monitoring the time-dependent morphological and structural response of a material to the alteration of process conditions. Complimentary to in situ TEM, which allows for highly localized, micro- and nano-scale analysis, in situ XRD can provide a similar type of information but at much larger scales, typically of the order of millimeters. In situ annealing XRD experiments performed using a calibrated thermal stage can give information on the overall structural evolution of a sample:

temperature-dependent nucleation, thermally-driven phase transitions or

crystallite size changes, phase composition changes, impurity formation, phase segregation, etc. Therefore, parallel in situ TEM and XRD experiments can provide detailed and complementary information about studied systems.


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 33

In this study, in situ XRD and TEM heating experiments of lithium iron phosphate (LiFePO4) precursors, prepared by a simple, low-cost sol-gel method, were carried out to investigate the evolution of the crystallization process and thermally-driven phase transitions in this material system under oxygen-rich and oxygen-deficiency conditions. In addition, a number of ex situ measurements were conducted, including Raman spectroscopy and electrochemical testing, to support the findings of these in situ studies and better understand this system. This work is focused on the optimization of synthesis conditions and emphasizes the importance of temperature-dependent investigations of solgel precursors.

Experimental Section Synthesis of LiFePO4 precursors. The analyzed samples were LiFePO4 precursors synthesized by a simple, low cost sol-gel method.25 First, iron powder (Fe), phosphoric acid (H3PO4) and lithium hydroxide monohydrate (LiOH·H2O) were dissolved in an aqueous solution of citric acid by continuous stirring. In the next step, sucrose was added as a carbon source, which is a typical procedure in such synthesis.26 Finally, the solution was evaporated at 60°C until a gel was obtained, which then was further oven-dried at 120°C. These prepared samples were pulverized in a homogenizer to obtain fine powders of LiFePO4 precursors. These powder precursors were used directly for in situ XRD measurements, whereas for in situ TEM study they were dispersed on dedicated 3.05 mmdiameter, 300 mesh gold grid-supported amorphous lacey carbon films.


4

Page 5 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Additionally, the samples were ex situ annealed between 400 and 900°C (with a 100°C step) in an inert gas atmosphere (N2) and measured using Raman and electrochemical techniques. Experimental Methods. Vacuum annealing (with a base pressure of about 10-7 Torr) of the precursor was conducted primarily in situ inside a 200 kV field-emission gun FEI Tecnai F20 TEM, using a furnace-type GATAN 628 Single Tilt Heating Holder. Various types of TEM-based measurements, including high resolution (HR) TEM, selected area electron diffraction (SAED), electron energy loss spectroscopy (EELS), and energy dispersive X-ray spectroscopy (EDX), were used to study thermally-driven changes in the structure and chemistry of the in situ heated precursors. In situ powder XRD annealing experiments were performed to monitor the crystallization process, identify phases present at different temperatures, characterize their crystal structures, and finally to compare with TEM data. XRD experiments were conducted using a Bruker Discovery D8 XRD system with a non-monochromatic Cu Kα radiation (λ = 0.15418 nm) and an Anton Paar DHS 1100 Domed Heating Stage. The XRD patterns were collected using the 2Theta coupled mode and a step size of 0.02°. Based on the signal-to-noise level in in situ XRD patterns, the lowest concentration of crystalline phase in the sample that could be detected was estimated at 15%. During in situ XRD and TEM, measurement samples were heated stepwise (using a 5oC/min heating rate) from room temperature to 800°C. At each heating step, after reaching the target temperature, a 5 min waiting time interval was included before starting the actual measurement. For comparison, the same precursors were also annealed ex situ at 700°C in a nitrogen atmosphere and examined with various methods, including XRD, SEM, TEM and EDX.


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 33

Additionally, a series of samples was prepared by annealing the same precursor at temperatures from 400 to 900°C (with 100°C steps) and measured by micro-Raman spectroscopy and electrochemical testing. The Raman analysis, which provided additional chemical and structural data (including chemical maps), was performed using a Renishaw in Via Raman Microscope, equipped with a Nd:YAG laser with an emission line of 532 nm. In order to avoid any possible laser-induced thermal decomposition of the samples, its power was significantly reduced for these measurements.25,27,28 The micro-Raman point maps were based on spectra recorded in the sample areas of 900 µm2 using the 0.5 µm step (3721 points), and the most intense peak from each of the phases observed in the spectra was mapped. The morphology of the ex situ annealed (700°C) samples was also analyzed using scanning electron microscopy (SEM) in a FEI Nova 600 microscope. The electrochemical performance of the lithium iron phosphate powders was studied in three-electrode Swagelok® - type cells with lithium metal as the counter and reference electrodes, a Celgard® 2402 polypropylene porous membrane as the separator, and 1M LiPF6 in a 50:50 v/v mixture of ethyl carbonate (EC) and dimethyl carbonate (DMC) as an electrolyte. All cells were prepared in an Ar-filled glove box with oxygen and moisture levels lower than 1 ppm. All cathodes were made by mixing 83 wt% active materials with 10 wt% of carbon additive and 7% of polivinylidene fluoride (PVDF) as a binder in Nmethyl-2-pyrrolidon (NMP). The slurry was mixed intensively for 24h, coated onto aluminum foil, and oven-dried overnight at 120°C. The charge/discharge measurements were carried out over a potential range between 3.2 V and 3.8 V with a multi-channel battery tester, the Atlas Solich 0965. For charge/discharge cycling, a current of 17 mA g-1 (0.1 C) was used based on the cathode (LFP) mass.


6

Page 7 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Results and discussion A summary of the characterization results from typical LiFePO4 precursors, produced through our sol-gel synthesis, before and after ex situ annealing at 700oC, is presented in Fig. S1-S3, Supporting Information. The study showed that the as-synthesized precursors had porous morphology and, as confirmed by both XRD and TEM measurements, were amorphous with no crystalline phases present (Fig. S1, Supporting Information). However, the ex situ annealing at 700oC resulted in full crystallization. The XRD measurements of such material showed a well-developed pattern in a pure LiFePO4 crystal structure (ICDD PDF 01-070-6684). The particles in these samples, as observed in TEM images, had an irregular and porous morphology. HRTEM images of formed crystallites reveled well-defined atomic planes with expected d-spacing values for LiFePO4 structure (in particular: d101 = 4.27 Å, Fig. S2, Supporting Information). TEM analysis and Raman spectra additionally indicated the presence of an amorphous carbon in these samples. These basic measurements were a starting point for a detailed in situ analysis of annealing temperature’s impact on the electrochemical performance of sol-gel synthesized LiFePO4 materials. In situ XRD annealing. In situ XRD heating experiments revealed that the synthesis process can take different routes depending on the annealing atmosphere. We observed that annealing in air with unrestricted oxygen access leads to the formation of two phases, namely Li3Fe2(PO4)3 and α-Fe2O3 (Fig. S4, Supporting Information). These two products are also observed by Raman spectroscopy after laser-induced thermal decomposition of LiFePO4 under unrestricted oxygen access conditions.25,27,28 Our in situ XRD annealing experiments in air showed that crystallization started at about 450°C, where the color of


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 33

Figure 1. In situ XRD heating of LiFePO4 in non-oxygen condition showing evolution of diffraction peaks.

the sample changed from green (a precursor color containing Fe2+ ions) to red (an indication of a residual α-Fe2O3 phase formation) which additionally confirmed the reaction.

Figure 2. In situ XRD heating of LiFePO4 in non-oxygen conditions from RT to 900°C and then cooling to RT.


8

Page 9 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In the case of oxygen-deficiency conditions, the nucleation started at around 425– 445°C, where reflections from the LiFePO4 phase (Pnma space group) began to appear (ICDD PDF # 01-070-6684). With increasing temperature these diffraction peaks became sharper and more intense, which suggested a systematic crystallite size increase (Fig. 1). At 500°C diffraction peaks were already 3 times stronger, and at 700°C nearly 5 times stronger than measured at 450°C. The complete evolution of the crystallization process is presented in Fig. 2 showing the most intense reflection from the (311) plane and a few minor reflections from the (301), (121) and (410) planes at higher temperatures. Weak peaks with intensities slightly above the noise level for reflections around 40° (i.e. (102), (221) and (401)) were seen above 475°C. The average crystallite sizes, calculated using the Scherrer equation29 from the (311) diffraction peak (which was the most intense one in the 2-Theta range measured) are listed as a function of annealing temperature in Table 1.

Table 1. Average crystallite sizes, calculated using the Scherrer equation from the (311) diffraction peak, as a function of annealing temperature;the last value was obtained after cooling the sample down to 25°C.


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 33

The data shows that the average crystallite size increased from about 14 nm at the beginning of the crystallization process (425°C) to 79 nm at 900°C and then further to 83 nm after cooling the sample to room temperature. Quantitative analysis of acquired diffraction patterns was employed to extract unit cell parameters of the LiFePO4 structure as a function of annealing temperature (Fig S5a, Supporting Information). Interestingly, the data was found to show the quadratic thermal expansion of unit cell volume with increasing temperature of annealing (Fig. S5b, Supporting Information). A similar effect has been previously observed for other olivine-type compounds.30 From our data, the thermal volume expansion coefficient was estimated at 0.01 Å3/°C. After cooling the material to 25°C the unit cell volume decreased to approximately 291.5 Å3 (a = 10.322 Å, b = 6.008 Å, c = 4. 699 Å), which is a typical room temperature value for defect-free LiFePO4.22,23 Long acquisition times for each temperature were infeasible, which limited the sensitivity of our in situ XRD measurements; any impurities or low-concentration minority phases which could possibly be present were not detected. Additional impurity analysis was performed using micro-Raman mapping, which is discussed later. In situ TEM annealing. In situ TEM heating measurements showed the change in the morphology

of

the

samples

as

well

as

the

amorphous-crystalline

transition.

Complementary studies were performed using SAED and EELS to identify crystalline phases, evaluate structural quality and morphology of the crystallites, and determine the type of carbon-based structures formed during annealing. In situ HRTEM analysis confirmed that the precursor was fully amorphous at the beginning of the annealing process and mesoporosity (with the average pore size of about 2 nm) was visible in the


10

Page 11 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. In situ TEM heating of a LiFePO4 precursor between 400 and 800°C. The first two columns show TEM images of the same grain; the third and fourth show HRTEM images, and the fifth shows SAED patterns in the same region. Notice a slight grain rotation at 800°C compared with 700°C and lower temperatures, which resulted in a change of the zone axis, a different SAED pattern, and different planes visible in HRTEM images.


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 33

Figure 4. HRTEM images taken at 450°C showing the beginning of the annealing related crystallite formation (crystal planes marked on pink) in the amorphous precursor grain. Images taken (a) just after reaching the set temperature and (b) 5 minutes later.

particles. The crystalline lattice fringes were initially observed between 450 and 500°C (Fig. 3), which was consistent with SAED analysis. At 450°C HRTEM images revealed that lattice fringes were already visible for some particles (d101 = 4.27 Å, Fig. 4), but others showed small clusters of arranged atoms

which is an early indication of

nucleation. The majority of clusters were, however, too small to give any significant diffraction signal; the SAED pattern recorder at this temperature still contained mostly diffuse amorphous rings (Fig. 3). Nevertheless, the nucleation point estimated from our in situ TEM investigation agrees with the one measured from the in situ XRD analysis. The SAED and HRTEM analyses showed that an increase of annealing temperature above the nucleation point between 500°C and 700°C resulted in the formation and growth of LiFePO4 crystallites (d211 = 3.01 Å). The SAED patterns at these temperatures contained slightly arced and multiplied spots, which suggested that the grains were textured


12

Page 13 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


polycrystallites. During further annealing, these grains arranged into bigger crystals (at 800°C) despite the randomness of their initial nucleation and the presence of polycrystallites at lower temperatures. Based on TEM images, with increasing annealing temperature the particles’ morphology became more porous, and their surfaces and shapes changed slightly (Fig. 3) (though the overall particle size remained nearly static above 800°C). This observed temperature dependent evolution of sample porosity was further confirmed using the Brunauer-Emmett-Teller analysis (Fig S6, Supporting Information).

Figure 5. EELS spectra of the carbon K-edge energy region showing the evolution of carbon structure during annealing.

EELS spectra of the carbon K-ionization edge showed that the chemical state of the carbon precursor changed significantly at temperatures lower than the crystallization of LiFePO4 (Fig 5). It was observed that the initial organic phase decomposed between 200 and 350°C, forming an amorphous carbon phase. Further carbon characterization was performed by ex situ Raman spectroscopy.


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 33

Figure 6. Raman spectra of the samples annealed between 400 and 900°C. Changes in the D and G peaks of carbon and growth of LiFePO4 peaks caused by annealing are visible.

Ex situ micro-Raman analysis. Micro-Raman spectroscopy, including mapping, was used to characterize six samples annealed between 400 and 900°C. This characterization focused on the LiFePO4 and carbon structures as well as the impurity phases. Figure 6 shows a comparison of spectra of samples annealed between 400 and 900°C. These spectra were normalized to the G peak of carbon in order to see the difference in intensities between LiFePO4 phases. While samples annealed at temperatures between 500 and 800°C showed only the most intense peak of the LiFePO4 phase, i.e. the peak located at 950 cm-1 (see enlarged spectra in Fig. S7, Supporting Information), the material annealed at 900°C had several sharp lines, most of them characteristic for LiFePO4 in an olivine crystal structure. The peak at 950 cm-1 (with the width of 7 cm-1) originated from a fully symmetric Agν1 mode (known as the breathing mode) of a tetrahedral PO4 group.25,31,32 Two weaker lines in the spectrum of LiFePO4 (widths: 10 and 8 cm-1) located at 997 and 1068 cm−1 were related to vibrational stretching modes of the PO4


14

Page 15 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


tetrahedron (ν3 modes).25,31,32 Additionally, there were three bands at 630, 590 and 573 cm−1 which corresponded to the bending modes of LiO6 and PO4. Finally, the broader bending mode of FeO6 and PO4 located at 442 cm−1 was also present in the 900oC spectrum. Consistent with the XRD and TEM studies, the measured Raman line widths strongly confirmed good structural quality of LiFePO4 formed at 900°C. Carbonaceous structure phonon modes with broad lines at 1342 cm−1 (D line) and 1599 cm−1 (G line) were observed for each sample.33,34 A few other unknown broad bands located before the D line and between the D and G peaks were also visible in the spectra; their intensities changed with the temperature of annealing. They might be related to the decomposition of the residual organic substrates (sucrose and citric acid), especially those bands located at about 1100 and 1450 cm-1 which are usually assigned to vibrations from various functional groups or C-H deformation.35,36 The widths of the D and G lines are about 150 and 100 cm−1, respectively, which proves that the carbon is certainly amorphous. The quality of the carbon slightly improved with increasing annealing temperature, as indicated by a decreasing ID/IG ratio trend (e.g. for samples annealed at 700, 800 and 900°C, the ID/IG ratio were 2.69, 2.66, 2.34, respectively).

Figure 7. Raman mapping of the samples annealed at 500, 700 and 900°C showing the formation of LiFePO4 crystallites (red) and decline of α-Fe2O3 impurity phase (green) during heating.


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 33

Raman mapping of the samples annealed at 500, 700 and 900°C provided additional information about the impurities present in the samples (Fig. 7). The impurity signal was already observed in the spectra of the 400°C sample and the peaks were assigned to an αFe2O3 structure (Fig. S7, Supporting Information).25,28 The micro-Raman map of the material annealed at 500°C showed a high concentration of the α-Fe2O3 phase (green spots) compared to trace amounts of the LiFePO4 phase (red spots). The α-Fe2O3 phase is probably in an amorphous state and hence not observed in the in situ XRD measurements. Even the most intense peak of the LiFePO4 structure was hardly visible in the spectra. The intensity of this peak increased for the sample annealed at 700°C, whereas the amount of the impurity phase of α-Fe2O3 decreased to the level of about 10% of the map area. The spectra set of the 900°C map showed sharp lines and a smooth distribution of high quality LiFePO4 crystallites. The impurity concentration decreased even further, to the level of about 0.5 % of the mapped area.

Figure 8. Initial charge/discharge curves at 0.1 C rate of LiFePO4 samples annealed at various temperatures.


16

Page 17 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Electrochemical testing. The initial charge/discharge curves at a current rate 0.1 C of samples synthesized between 500 and 900oC are shown in Fig. 8. The specific capacities of all these materials revealed during first charge were much higher than during discharge, probably due to the formation of interlayers at the LFP/electrolyte and the LFP/current collector interface and/or oxidation of impurities.37 This resulted in a rather poor coulombic efficiency of the first cycle of about 65%. The indication of these processes could be also noticed in the charge profiles of the first oxidation, where no stable plateau was observed. The deintercalation of lithium ions from the LFP structure occurred at the stable potential plateau of about 3.4 V, indicating a two-phase electrochemical reaction between LiFePO4 and FePO4. It is worth mentioning that values of the potential hysteresis calculated in the middle of the intercalation/deintercalation processes (summarized in Table 2) depended on the material, which indicated various overall cell resistances. A minimal difference in potential plateau between the charge and discharge steps was observed for samples synthesized at 800°C (70 mV at the first cycle and 40 mV at subsequent cycles), which corresponded to a total cell resistance of 2.6 kΩ.

Table 2. Potential hysteresis between charge and discharge profiles and calculated overall cell resistances of samples synthesized at various temperatures.


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 33

Fig. S8, Supporting Information, shows discharge specific capacities at a current rate of 0.1 C of the samples synthesized at various temperatures. The material annealed at 400°C was inactive at the examined potential range, probably due to an incomplete crystallization process and/or a high amount of impurities. The specific discharge capacities were 55, 65, 75 and 80 mAh g-1 for samples synthesized at 500, 600, 700 and 800°C, respectively. It can be observed that the specific capacity of material annealed at 900°C decreased to 27 mAh g-1, which could be the result of LFP grain agglomeration. It is also worth mentioning that the specific capacity of samples annealed between 500 and 800°C slightly increased during the first five cycles. This phenomenon has also been observed by Kamarulzaman et al. in lithium manganese oxide LiMn2O4, attributed to the conditioning effects of the positive electrode during the charge–discharge processes.38 Discussion It was claimed by Chung et al. that rapid crystal growth can cause a disorder of the cations in the LiFePO4 structure.20 In this study we obtained single crystals with a good crystal quality. These crystallites showed defect-free unit cell parameters. The unchanged particle size after annealing indicated that the crystallite size growth took place internally within the grain. Due to technical limitations, we could not perform TEM annealing at temperatures higher than 800°C, so grain coarsening above this temperature could not be observed directly. It is important to note that in the in situ TEM set-up, a temperature gradient is very probable, because of the limited heat transfer under vacuum conditions. This could be responsible for the slight differences between starting crystallization temperatures estimates based on in-situ TEM and XRD experiments.


18

Page 19 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Although the specific capacity values measured in our study diverged from the best ones reported in literature39,40 (probably due to a lack of efficient carbon coating layers and higher grain sizes), our work demonstrated that a combination of simultaneous in-situ and ex-situ measurements could be an effective way to study details of the synthesis process to better understand the relationships between the structure and electrochemical properties of electrode materials. Particularly, the study showed that the electrochemical properties and the structure of synthesized materials were strongly affected by annealing temperature. At low-temperature treatment, an incomplete crystallization process resulted in a lack of electrochemical activity of the sample. The powders annealed at temperatures higher than 400°C revealed specific capacities between 55 and 80 mAh g -1 depending on crystallization temperature. An increase of the synthesis temperature resulted in a decrease of impurity concentration (mainly α-Fe2O3 phases, which has also been reported previously for synthesis of LiFePO4 using other routes, such as hydrothermal method with LiOH and citric acid as substrates)23,41 and a higher level of crystallization, which resulted in better electrochemical performance. LiFePO4 annealed at 900°C revealed a severe decrease of specific capacity (down to about 30 mAh g-1), mainly due to an agglomeration of active particles42 and an excessive increase in crystallite size resulting in a lowering of the ionic conductivity of the material and a much higher resistivity of the electrochemical cell (Fig. 8, S8-S11, Supporting Information).43,44 As shown in SEM and STEM images (Fig. S10, Supporting Information) the porosity of the material and carbon morphology changed for samples annealed at higher temperatures. The carbon porosity developed while annealing at higher temperatures, whereas the porosity of LiFePO4 crystallites and grains significantly decreased above 800°C. LiFePO4 crystallites agglomerated and


19


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 33

separated from carbon structure; their surfaces became smooth, leaving no access for electrolytes. Conclusions A combination of in situ XRD and TEM heating experiments was employed to study the sol-gel-based synthesis of LiFePO4. The results demonstrated that simultaneous application of these two in situ techniques can provide details, at different length scales, of structural and morphological changes which occur in a material during various stages of the synthesis process, including the crystallization (nucleation and growth) of LiFePO4. Supporting ex-situ Raman spectroscopy and electrochemical measurements of samples from the same precursors agreed with the structural evolution revealed by in-situ XRD and TEM experiments. The increased crystallization and improved structural quality of LiFePO4 with increasing annealing temperature, observed by XRD, TEM and also suggested by Raman spectroscopy, resulted in improved electrochemical properties (with the best electrochemical performance obtained from material synthesized at 800°C). Further increase of the synthesis temperature to 900oC resulted in a decrease of the electrochemical properties, due to the formation of large LiFePO4 particles and limited ionic transport. Our study can serve as an example showing the importance of a combination of in situ experiments and electrochemical measurements to the optimization of synthesis conditions and efficiency improvements in lithium ion battery technology. A demonstrated experimental approach can be applied to any electrochemical system.


20

Page 21 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Supporting Information. TEM, SEM, EELS and XRD ex situ data from the as-prepared precursor sample and from the sample annealed at 700°C. Additional XRD, Raman, BET, STEM, SEM, TEM and electrochemical results from a series of samples annealed ex-situ between 400 and 900°C. These materials are available free of charge via the Internet at http://pubs.acs.org. Corresponding Authors *

Corresponding authors: [email protected], [email protected]

Author Contributions D.A.Z, J.B.J., and B.H. conceived and conducted experiments, wrote the manuscript and supplementary information text, and prepared Figures. D.A.Z. performed synthesis and preparation of samples. J.B.J. performed TEM, EELS, and SAED, D.A.Z. performed XRD, SEM and Raman, D.A.Z. and B.H. performed electrochemical analysis of the samples. All authors participated in discussions as well as reviewed and commented on the manuscript. All authors have given approval to the final version of the manuscript.

Funding Sources US

National

Science

Foundation-

EPSCoR

Program:

Sub-Award No.

3048111570-15-016

National Centre for Research and Development (NCBiR): PBS1 research grant, Contract No. PBS1/A1/4/2012.

Acknowledgments. The work of DZ was supported by the Foundation for Polish Science International PhD Projects Programme co-financed by the EU European Regional Development Fund and The Management Board of the Mazowieckie Voivodeship PhD scholarship programme of the Operational


21


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 33

Programme Human Capital 2007-2013 co-financed by EU European Social Fund. This work was partially supported by the US National Science Foundation EPSCoR Program (sub-award no. 3048111570-15-016) and the Polish National Centre for Research and Development (NCBiR) through the research grant PBS1 (contract no. PBS1/A1/4/2012).

References (1) (2) (3) (4) (5) (6)

(7) (8) (9) (10) (11) (12)

(13) (14) (15)

Scrosati, B.; Garche, J. J. Power Sources 2010, 195, 2419. Fergus, J. W. J. Power Sources 2010, 195, 939. Padhi, A. K.; Nanjundaswamy, K. S.; Goodenough, J. B. J. Electrochem. Soc. 1997, 144, 1188. Whittingham, M. Chem. Rev. 2004, 10, 4271. Ramana, C. V.; Mauger, A.; Gendron, F.; Julien, C. M.; Zaghib, K. J. Power Sources 2009, 187, 555. Orikasa, Y.; Maeda, T.; Koyama, Y.; Minato, T.; Murayama, H.; Fukuda, K.; Tanida, H.; Arai, H.; Matsubara, E.; Uchimoto, Y.; Ogumi, Z. J. Electrochem. Soc. 2013, 160, A3061. Niu, J.; Kushima, A.; Qian, X.; Qi, L.; Xiang, K.; Chiang, Y.-M.; Li, J. Nano Lett. 2014, 14, 4005. Wang, X.-J.; Chen, H.-Y.; Yu, X.; Wu, L.; Nam, K.-W.; Bai, J.; H. Li,; Huang, X.; Yang, X.-Q. Chem. Commun. 2011, 47, 7170. Chen, Y.-C.; Chen, J.-M.; Hsu, C.-H.; Lee, J.-F.; Yeh, J.-W.; Shih, H. C. Solid State Ionics 2009, 180, 1215. Wang, X.-J.; Jaye, C.; Nam, K.-W.; Zhang, B.; Chen, H.-Y.; Bai, J.; Li, H.; Huang, X.; Fischer, D. A.; Yang, X.-Q. J. Mater. Chem. 2011, 21, 11406. Yu, X.; Wang, Q.; Zhou, Y.; Li, H.; Yang, X.-Q.; Nam, K.-W.; Ehrlich, S. N.; Khalid, S.; Meng, Y. S. Chem. Commun. 2012, 48, 11537. Leriche, J. B.; Hamelet, S.; Shu, J.; Morcrette, M.; Masquelier, C.; Ouvrard, G.; Zerrouki, M.; Soudan, P.; Belin, S.; Elkaïm, E.; Baudelet, F. J. Electrochem. Soc. 2010, 157, A606. Orikasa, Y.; Maeda, T.; Koyama, Y.; Murayama, H.; Fukuda, K.; Tanida, H.; Arai, H.; Matsubara, E.; Uchimoto, Y.; Ogumi, Z. J. Am. Chem. Soc. 2013, 135, 5497. Orikasa, Y.; Maeda, T.; Koyama, Y.; Murayama, H.; Fukuda, K.; Tanida, H.; Arai, H.; Matsubara, E.; Uchimoto, Y.; Ogumi, Z. Chem. Mater. 2013, 25, 1032. Wang, C. M.; Xu, W.; Liu, J.; Choi, D. W.; Arey, B.; Saraf, L. V.; Zhang, J. G.; Yang, Z. G.; Thevuthasan, S.; Baer, D. R.; Salmon, N. J. Mater. Res. 2010, 25, 1541.


22

Page 23 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(16) Gu, L.; Zhu, C.; Li, H.; Yu, Y.; Li, C.; Tsukimoto, S.; Maier, J.; Ikuhara, Y. J. Am. Chem. Soc. 2011, 133, 4661. (17) Zhu, Y.; Wang, J. W.; Liu, Y.; Liu, X.; Kushima, A.; Liu, Y.; Xu, Y.; Mao, S. X.; Li, J.; Wang, C.; Huang, J. Y. Adv. Mater. 2013, 25, 5461. (18) Chung, S.-Y.; Kim, Y.-M.; Kim, J.-G.; Kim, Y.-J. Nature Physics 2009, 5, 68. (19) Trudeau, M. L.; Laul, D.; Veillette, R.; Serventi, A. M.; Mauger, A.; Julien, C. M.; Zaghib, K. J. Power Sources 2011, 196, 7383. (20) Chung, S.-Y.; Kim, Y.-M.; Lee, S.; Ho Oh, S.; Kim, J.-G.; Choi, S.-Y.; Kim, Y.-J.; Kang, S.-J. L. Nano Lett. 2012, 12, 3068. (21) Chung, S.-Y.; Kim, Y.-M.; Choi, S.-Y.; Kim, J.-G. J. Am. Chem. Soc. 2013, 135, 7811. (22) Chen, J.; Bai, J.; Chen, H.; Graetz, J. J. Phys. Chem. Lett. 2011, 2, 1874. (23) Chen, J.; Graetz, J. ACS Appl. Mater. Interfaces 2011, 3, 1380. (24) Chen, Z.; Ren, Y.; Qin, Y.; Wu, H.; Ma, S.; Ren, J.; He, X.; Sun, Y.-K.; Amine, K. J. Mater. Chem. 2011, 21, 5604. (25) Ziolkowska, D.; Korona, K. P.; Hamankiewicz, B.; Wu, S.-H.; Chen, M.-S.; Jasinski, J. B.; Kaminska, M.; Czerwinski, A. Electrochim. Acta 2013, 108, 532. (26) Wu, S.-H.; Hsiao, K.-M.; Liu, W.-R. J. Power Sources 2005, 146, 550. (27) Burba, C. M.; Palmer, J. M.; Holinsworth, B. S.; J. Raman Spectrosc. 2009, 40, 225. (28) Markevich, E.; Sharabi, R.; Haik, O.; Borgel, V.; Salitra, G.; Aurbach, D. J. Power Sources 2011, 196, 6433. (29) Patterson, A.L. Phys. Rev. 1939, 56, 978. (30) Hazen, R. M. Am. Min. 1977, 62, 286. (31) Paques-Ledent, M. T.; Tarte, P. Spectrochim. Acta 1973, 29A, 1007. (32) Korona, K. P.; Papierska, J.; Kaminska, M.; Witowski, A.; Michalska, M.; Lipinska, L. Mater. Chem. Phys. 2011, 127, 391. (33) Tuinstra, F.; Koenig J.L., J. Chem. Phys. 1970, 53, 1126. (34) Chu, P. K.; Li, L. Mater. Chem. Phys. 2006, 96, 253. (35) De Gelder, J.; De Gussem, K.; Vandenabeele, P.; Moens, L. J. Raman Spectrosc. 2007, 38, 1133. (36) Brizuela, A. B.; Bichara, L. C.; Romano, E.; Yurquina, A.; Locatelli, S.; Brandán, S. A. Carbohydr. Res. 2012, 361, 212. (37) Gellert, M.; Gries, K. I.; Zakel, J.; Ott, A.; Spannenberger, S.; Yada, C.; Rosciano, F.; Volz, K.; Roling, B. Electrochim. Acta 2014, 133, 146. (38) Kamarulzaman, N.; Yusoff, R.; Kamarudin, N.; Shaari, N. H.; Abdul Aziz, N. A.; Bustamb, M. A.; Blagojevic, N.; Elcombe, M.; Blackford, M.; Avdeev, M.; Arof, A. K. J. Power Sources 2009, 188, 274. (39) Ferrari, S.; Lassarote Lavall, R.; Capsoni, D.; Quartarone, E.; Magistris, A.; Mustarelli, P.; Canton, P. J. Phys. Chem. C 2010, 114, 12598. (40) Yamada, A.; Chung, S. C.; Hinokuma, K. J. Electrochem. Soc. 2001, 148, A224.


23


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 33

(41) Shiraishi, K.; Dokko, K.; Kanamura, K. J. Power Sources 2005, 146, 555. (42) Scaccia, S.; Carewska, M.; Wisniewski, P.; Prosini, P. P. Mater. Res. Bull. 2003, 38, 1155. (43) Gaberscek, M.; Dominko, R.; Jamnik, J. Electrochem. Commun. 2007, 9, 2778. (44) Yang, X.; Liu, D.; Xu, X.; He, X.; Xie, J. CrystEngComm 2013, 15, 10648.


24

Page 25 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


For Table of Contents Use Only

In Situ XRD and TEM Studies of Sol-Gel-based Synthesis of LiFePO4 Dominika A. Ziolkowskaa,*, Jacek B. Jasinskib,*, Bartosz Hamankiewiczc, Krzysztof P. Koronaa, She-Huang Wud and Andrzej Czerwinskic

Parallel in situ TEM and XRD studies on LiFePO4 precursors obtained by sol-gel method are performed to study evolution of the material during synthesis annealing, which is a crucial stage in preparing rechargeable cathode materials. An exemplary optimization of the final sol-gel step is shown by comparing in situ heating results with electrochemical performance of materials annealed at different temperatures.


25


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figures

Figure 1. In situ XRD heating of LiFePO4 in non-oxygen condition showing evolution of diffraction peaks.


Page 26 of 33

Page 27 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. In situ XRD heating of LiFePO4 in non-oxygen conditions from RT to 900°C and then cooling to RT.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. In situ TEM heating of a LiFePO4 precursor between 400 and 800°C. The first two columns show TEM images of the same grain; the third and fourth show HRTEM images, and the fifth shows SAED patterns in the same region. Notice a slight grain rotation at 800°C compared with 700°C and lower temperatures, which resulted in a change of the zone axis, a different SAED pattern, and different planes visible in HRTEM images.


Page 28 of 33

Page 29 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. HRTEM images taken at 450°C showing the beginning of the annealing related crystallite formation (crystal planes marked on pink) in the amorphous precursor grain. Images taken (a) just after reaching the set temperature and (b) 5 minutes later.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. EELS spectra of the carbon K-edge energy region showing the evolution of carbon structure during annealing.


Page 30 of 33

Page 31 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 6. Raman spectra of the samples annealed between 400 and 900°C. Changes in the D and G peaks of carbon and growth of LiFePO4 peaks caused by annealing are visible.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 7. Raman mapping of the samples annealed at 500, 700 and 900°C showing the formation of LiFePO4 crystallites (red) and decline of α-Fe2O3 impurity phase (green) during heating.


Page 32 of 33

Page 33 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 8. Initial charge/discharge curves at 0.1 C rate of LiFePO4 samples annealed at various temperatures.


In Situ XRD and TEM Studies of Sol-Gel-Based ... - ACS Publications

Recommend Documents