Significantly enhanced electrochemical performance of lithium titanate anode for lithium ion battery by the hybrid of nitrogen and sulfur co-doped graphene quantum dots

Accepted Manuscript Title: Significantly enhanced electrochemical performance of lithium titanate anode for lithium ion battery by the hybrid of nitrogen and sulfur co-doped graphene quantum dots Author: Li Ruiyi Jiang Yuanyuan Zhou Xiaoyan Li Zaijun Gu Zhiguo Wang Guangli Liu Junkang PII: DOI: Reference:

S0013-4686(15)30268-1 http://dx.doi.org/doi:10.1016/j.electacta.2015.08.018 EA 25486

To appear in:

Electrochimica Acta

Received date: Revised date: Accepted date:

29-6-2015 3-8-2015 4-8-2015

Please cite this article as: Li Ruiyi, Jiang Yuanyuan, Zhou Xiaoyan, Li Zaijun, Gu Zhiguo, Wang Guangli, Liu Junkang, Significantly enhanced electrochemical performance of lithium titanate anode for lithium ion battery by the hybrid of nitrogen and sulfur co-doped graphene quantum dots, Electrochimica Acta http://dx.doi.org/10.1016/j.electacta.2015.08.018 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Significantly enhanced electrochemical performance of lithium titanate anode for lithium ion battery by the hybrid of nitrogen and sulfur co-doped graphene quantum dots Li Ruiyia, Jiang Yuanyuana, Zhou Xiaoyana, Li Zaijun*a,b, Gu Zhiguoa, Wang Guanglia and Liu Junkanga a:School of

Chemical and Material Engineering, Jiangnan University, Wuxi

214122, China b:Key

Laboratory of Food Colloids and Biotechnology, Ministry of Education,

Wuxi 214122, China

Graphical abstract The study reported a facile synthesis of Li4Ti5O12/nitrogen and sulfur co-doped graphene

quantum dots (LTO/N,S-GQDs). The

unique

architecture and the introduction of N,S-GQDs create both ultrafast electron

transfer

and

electrolyte

transport.

The

as-prepared

LTO/N,S-GQDs anode provides prominent advantage of specific capacity, high-rate performance and cycle stability.

Highlights *Corresponding author.

Tel.:13912371144 Fax.: +86 0510 85811863. E-mail address: [email protected].

► We reported a new lithium titanate/nitrogen and sulfur co-doped graphene quantum dots hybrid ► The synthesis creates a crystalline interconnected porous framework composed of nanoscale LTO ► The unique architecture achieves to maximize the rate performance and enhance the power density ► Introduction of N,S-GQDs greatly enhances the electron transfer and the storage lithium capacity ► The hybrid anode provides an excellent electrochemical performance for lithium-ion batteries ABSTRACT The paper reported a facile synthesis of lithium titanate/nitrogen and sulfur co-doped graphene quantum dots(LTO/N,S-GQDs). Tetrabutyl titanate was dissolved in tertbutanol and heated to refluxing state by microwave irradiation. Then, lithium acetate was added into the mixed solution to produce LTO precursor. The precursor was hybridized with N,S-GQDs in ethanol. Followed by drying and thermal annealing at 500°C in Ar/H2 to obtain LTO/N,S-GQDs. The synthesis creates fully crystalline interconnected porous framework composed of nanoscale LTO crystals. The unique architecture achieves to maximize the high-rate performance and enhance the power density. More importantly, the introduction of N,S-GQDs don’t almost influence on the electrolyte transport, but greatly improve the electron transfer and the storage lithium capacity. The LTO/N,S-GQDs anode exhibits remarkably enhanced electrochemical

performance for lithium ion battery. The specific discharge capacity is 254.2 mAh g-1 at 0.1C and 126.5 mAh g-1 at 10C. The capacity remains 96.9% at least after 2000 cycles at 2C. The battery performance is significantly better than that of pure LTO electrode and LTO/graphene electrode. Keywords:

Lithium

titanate;

graphene

quantum

dots;

electrochemical

performance; lithium ion battery 1. Introduction Lithium ion battery is popular electrochemical devices since its first commercialization by Sony Corporation in 1990s. It has been widely used in various portable electronic devices, including notebook personal computer, mobile phone, tablet and digital camera, due to its merits in terms of high operating voltage, high energy density, low self-discharge and the absence of memory effects. Its applications have been extended to electric vehicles and hybrid electrical vehicles to meet the environmental concerns [1]. As battery performance of lithium ion battery seriously depends on the electrode materials, the development of new electrode materials with high-performance has become the focus of lithium battery industry [2]. To date, graphite still is a main anode

material for commercial lithium ion batteries. Graphite is rich in resources, low cost and high electronic conductivity, but it has low lithium

ion

diffusion

coefficient,

restricting

its

applications

in

high-power lithium ion battery. In addition, low operating potential below 0.2 V (vs. Li/Li+) may result in the growth of lithium dendrites on the anode surface in the overcharged state [3]. In the last decade, great effort has been paid to search for alternative anode candidates instead of conventional graphite electrode [4]. Currently, lithium titanate (LTO) is considered as the most promising one, because of its excellent safety characteristic and long lifetime. LTO anode provides many advantages such as excellent lithium ion insertion and extraction reversibility, negligible volume change and structural change during the charge and discharge process, and a flat potential plateau. The vital drawback for LTO is its intrinsic insulating property, leading to severe polarization discharged at high current density. This will bring a poor high-rate performance and limits its applications in high-performance lithium ion battery [5]. Several strategies have been successfully attempted to improve the electrochemical properties of LTO in the resent years [6]. The first is to fabricate nanoscale LTO crystals. The nanosizing

effectively enhances

the high-rate performance of lithium ion battery via shortening diffusion distance of lithium ions in the electrode materials. Fattakhova-Rohlfing et al. synthesized fully crystalline interconnected porous frameworks composed of ultra small LTO nanocrystals of a few nanometers in size [7]. These framework feature a gravimetric capacity of about 175 mAh g-1 at rates of 1-50 C and can deliver up to 73% of their maximum capacity at unprecedented high rates of up to 800C without the deterioration up to 1000 cycles. However, the nanosizing inevitably leads to reduce the tap density, resulting in a low volumetric specific capacitance. This will greatly limit its applications in high-energy lithium ion battery. In addition, the production of nanoscale LTO crystals often requires a complex process, multi steps and serious control of the reaction conditions, which results in an obvious increase of product cost. The second is to introduce metal ions or atoms into LTO crystals. The doping can remarkably enhance the high-rate performance of lithium ion battery, owing to an improved electronic conductivity [8-11]. Czerwinski et al. developed a solid phase synthesis of the doping Ag into LTO [12]. The doped Ag atoms are distributed on the surface of LTO crystal as a conductive network, so as to enhance the high-rate capability. However, the doping may destroy the integrity of LTO crystal and result in the loss

of zero strain characteristic to some extent. This will greatly shorten the battery life. The third is to hybrid with other conductive materials to form LTO-based composite [13]. The composite shows a better rate capability than pure phase LTO, owing to a better electronic conductivity [14]. Recently, graphene as a new type of carbon materials has becomingly received great concern for the LTO modification. Graphene nanosheets are 2D macromolecular carbon material with remarkable electronic conductivity, large specific surface area and good mechanical property. The characteristics make it become an ideal conductive additive and structural support for LTO crystals. The investigation has acquired great achievement through mixing, coating or loading graphene on the LTO crystals. Chen et al. do well works in the mesoporous LTO grown on the reduced graphene oxide and the discharge capacity is 132 mAh g-1 at 40 C [15]. As graphene sheets are in the size of several microns to tens of microns, the coating such a large sheet on the surface of LTO crystals will hinder the entry of lithium ions into LTO crystal, leading to a remarkable decrease of the specific capacity. To resolve the problem, researchers had to convert 2D graphene into 3D graphene with special space structure or activated graphene with rich of mesopores for providing channels of the electrolyte transport. However, the building

3D graphene or activated graphene often requires a complex and time-consuming process. In addition, the use of large amounts of graphene also increases the cost, which are not in favour of the commercial process. More recently, graphene quantum dots (GQDs), single- or few-layer graphene with a tiny size of only several nanometers, have interesting optical properties due to tunable size and surface chemistry [16]. GQDs stand for a new type of QDs with the unique properties associated with both graphene and QDs, and have shown their value-added function in light emitting diode [17], supercapacitor [18], oxygen reduction reaction [19], solar cell [20] and sensors [21]. To the best of our knowledge, few report refers to the applications of GQDs in lithium ion battery. In the study, we focus a facile synthesis of lithium titanate/nitrogen and sulfur co-doped graphene quantum dots (LTO/N,S-GQDs). The as-prepared hybrid gives fully crystalline interconnected porous framework

composed

of

nanoscale

LTO

crystals.

The

unique

architecture and the introduction of N,S-GQDs create both ultrafast electron transfer and electrolyte transport and an enhanced specific capacity. The LTO/N,S-GQDs anode exhibits a prominent advantage of specific capacity, high-rate performance and cycle stability.

2. Experimental 2.1. N,S-GQDs preparation N,S-GQDs were prepared by thermal treatment of molecular organic salts with the mixed carbon source and the surface modifier in the single precursor [16]. In a typical preparation procedure, the mixture of citric acid (100 mmol) and L-cysteine (90 mmol) was dissolved in the deionized water (50 ml). Then, it was evaporated at 80 ºC until dry. The resulting thick syrup was hydrothermally heated in the Teflon-equipped stainless-steel autoclave at 220 ºC for 4 h and a heating rate of 10ºC min-1. The collected black syrup sample was neutralized with 1 M NaOH solution to pH 7 and finally dried by freeze drying to obtain a solid N,S-GQDs product. In addition, a common GQD was synthesized by using the same procedure unless no addition of L-cysteine. 2.2. LTO/N,S-GQDs synthesis Synthesis of LTO/N,S-GQDs includes the preparation of LTO precursor (p-LTO) and hybrid of p-LTO with N,S-GQDs and LTO/N,S-GQDs. In a typical procedure, 20 g of tetrabutyl titanate (TBT) were dissolved in 150 ml of tertbutanol (TBA). The mixed solution was transferred into a HWL07-3 microwave reactor and then heated to refluxing state using

microwave irradiation (300 W). Followed by adding the lithium acetate solution dissolved 4.9 g of lithium acetate (LiAc) in 20 ml of the deionized water into the mixed solution to form p-LTO. After 20 min, the solvent was removed from the reaction system by distillation. The collected p-LTO was dried at 200℃ for 2 h. After that, the p-LTO was mixed with the N,S-GQDs solution dissolved 0.4 g of N,S-GQDs in 50 ml of ethanol with the help of ultrasonic wave. Followed by drying and thermal annealing in Ar/H2 (95:5) atmosphere at 500°C with the temperate ramp rate of 10℃ min-1 for 12 h to obtain the LTO/N,S-GQDs product. To study on the effect of N,S-GQDs on the electrochemical property, LTO/GQDs or LTO-c) was prepared by the same procedure unless the use of GQDs instead of N,S-GQDs or no addition of N,S-GQDs. To compare the effect of graphene and N,S-GQDs on the battery performance of LTO, another hybrid, LTO/graphene (LTO-G), was also fabricated by the same procedure except for use of the same amounts of graphene oxide instead of N,S-GQDs. 2.3. Material characterization Scanning electron microscope (SEM) was performed using HITACHI S4800. Transmission electron microscope (TEM) was performed by a

JEOL 2010. X-ray diffraction (XRD) was measured on the D8 Advance with a Cu Kα radiation. Raman measurements were carried out using a InVia

laser

micro-Raman

spectrometer.

X-ray

photoelectron

spectroscopy (XPS) was performed by PHI 5700 using Al KR radiation. Fluorescence lifetime intensity decay of LTO/GQD and N,S-GQDs in aqueous solution was measured on the FLS 920 steady state and transient state fluorescence spectrometer (Edinburgh Instruments, England) with excitation at 370 nm and emission at 452 nm. 2.4. Electrochemical measurements Electrochemical properties of LTO materials were evaluated using 2016 coin cells. The active materials were mixed with super P conductive carbon and polyvinylidene fluoride (PVDF, Sigma-Aldrich) at weight ratio of 8:1:1 in N-methylpyrrolidone (NMP, Sigma-Aldrich) solvent to form uniform slurries, which were then coated on copper foils. The loading density of active materials was about 2.0 mg cm-2. Subsequently dried in a vacuum oven at 120°C overnight and rolled by using a rolling machine, these working electrodes were incorporated into 2016 coin cells, in which Li foils were serviced as the counter and reference electrode, Celgard 2400 as the separator, and a mixed solvent of

ethylene carbonate, dimethyl carbonate and diethylene carbonate (1:1:1) containing of 1 M LiPF6 as the electrolyte. The assembly process was conducted in an argon-filled glove box having O2 and H2 O contents below 0.1 ppm. Discharge-charge tests were performed at a potential range of 0.5~3.0 V (vs. Li/Li+) on a CT2001A LAND battery test system. All discharge-charge rates were denoted using the C-rate where 175 mA g-1 was assigned to be the current density of 1 C based on the theoretical capacity of LTO (175 mAh g -1). Cyclic voltammograms (CV) were performed in a CHI 660D electrochemical workstation over the potential range of 0.5~3.0 V at a scanning rate of 1 mV s -1 . Electrochemical impedance spectroscopy (EIS) measurements were carried out at 1.55 V on the coin cells using a CHI 660D electrochemical workstation. A potential amplitude of ±5 mV and a frequency range of 0.01 to 10 5 Hz were adopted. Before the CV and EIS measurements, the cells were cycled two times at 0.1C and subsequently equilibrated for 5 h at bias potential of 1.55V (vs. Li/Li+). 3. Results and Discussion 3.1. LTO/N,S-GQDs synthesis

Synthesis of LTO/N,S-GQDs includes three assemble processes, i.e. the preparation of p-LTO and hybrid of p-LTO with N,S-GQDs and LTO/N,S-GQDs (shown in Fig.1). First, TBT was dissolved in TBA and heated to refluxing state by microwave radiation. Followed by adding LiAc into the mixed solution to form p-LTO. In the step, the microwave radiation was used to reduce the reaction time. The microwave radiation make polar molecules produce strongly opposite movement and collision, which will accelerate the reaction rate. The result shows that the reaction can complete within 15 min. The time is less than conventional solvothermal synthesis [22]. In the study, TBA was employed as a novel medium for adjusting morphologies and particle sizes of p-LTO. Due to unique spatial structure and physicochemical properties, a large amount of w/o type of nanoemulsions occurred in the TBA/H2 O system [23], which would provide the limited space for forming small p-LTO particles. The structure-oriented and confinement effects will lead to form small and homogeneous nanostructures [7]. The result shows that the as-prepared p-LTO particles were in spherical shape with ultra small size (shown in Fig.s1). Further, these primitive p-LTO particles were reunited to form relatively big spherical agglomerates by the second agglomeration. This is important to form

fully crystalline interconnected porous framework composed of ultra small LTO crystals in the following step. Next, the p-LTO was hybridized with the N,S-GQDs in ethanol with the help of ultrasonic wave and dry to form the p-LTO/N,S-GQDs. To avoid further agglomeration, an ethanol solution was used as the dispersion medium for the hybridization of p-LTO with N,S-GQDs in the step. Because of the existence of rich hydrophilic groups such as hydroxyl and carboxyl groups, the N,S-GQDs are easy to dissolve in ethanol to form a homogeneous solution. The characteristic make N,S-GQDs can fill up all corners of the above agglomerates. Fig.s2 presents the photographs of p-LTO before and after the hybridized N,S-GQDs under the visible light and UV light radiation. It can be seen that the p-LTO sample itself has no obvious fluorescence emission under the visible light and UV light radiation. However, the p-LTO sample after hybridized N,S-GQDs exhibits a strong and uniform fluorescence under the UV light, indicating that the N,S-GQDs have been successfully modified on the surface of p-LTO particles. Finally, the p-LTO/N,S-GQDs was annealed at 500°C in Ar/H2 atomsphere to form LTO/N,S-GQDs. As a rule, the calcination temperature is the most important factor to influence on the crystallinity

and crystal size. The application of a relatively high temperature results in well-crystallized crystal with a relatively big crystal size. However, the electronic conductivity significantly decreases as the crystal size builds up. For the reason, researchers have to adopt a relatively low calcination temperature to obtain a high electrochemical performance [24]. There are still great amount of challenges to face for achieving both high volumetric specific capacitance and rate capacitance of LTO anode materials. In the study, we attempted to fabricate fully crystalline interconnected porous framework composed of small LTO crystals in order to make full use of the advantages of small crystal and large crystal. The assembly of nanoscale building blocks into interconnected porous frameworks is to maximize the rate performance and enhance the power density. Nanoscaling will increase the interface leading to enhanced charge transfer, and shortens the ion/electron diffusion pathways by decreasing the grain size of the bulk material. Our investigation demonstrated that the use of less than 400°C for the thermal annealing leads to the decomposition of spherical agglomerates composed of pristine p-LTO particles. The resulting LTO crystals show a particle size that is very close to that of the pristine p-LTO particles. If the temperature is higher than 600°C, the crystal size will rapidly

increase and eventually reach to several micron size at 800°C. When the temperature of about 500°C and fast heating rate (10°C min-1) were used for the thermal annealing, several p-LTO particles were naturally combined to form a relatively big LTO crystal. At the same, the most of N,S-GQDs were embedded into the interior of LTO crystal and bring a great improvement of the electronic conductivity. Interestingly, many nanoscale

LTO

crystals

are

constructed

into

fully

crystalline

interconnected porous framework with micron size due to the melting of the small LTO crystal edge. In addition, the most of hydrophilic groups on the surface of N,S-GQDs will be stripped during the heating. Since the hydrophilic groups of N,S-GQDs can combine with lithium ion in the electrolyte to produce an irreversible capacity, the reduction is very important to improve the cycle stability of the LTO/N,S-GQDs anode for lithium in battery. 3.2. Structure characterization Morphology and structure of the as-prepared LTO/N,S-GQDs were characterized by SEM and TEM technologies. Fig.2 shows that the LTO/Ni,S-GQDs is composed of the spherical agglomerates with the particle size of 1-2 µm (Fig.2a and c). The each of agglomerate consists of large number of nanoscale LTO crystals with the particle size of 30-60

nm (Fig.2b and d). The assembly of nanoscale building blocks into porous frameworks is to maximize the rate performance and enhance the power density. The nanoscaling will increase the interface, leading to an enhanced charge transfer and shortens the ion/electron diffusion pathways. Further, the enlarged TEM image demonstrates the existence of many N,S-GQDs inside or on the surface of LTO crystals. The size of N,S-GQDs is distributed in the range from 4 to 7 nm, with an average size of 5 nm. Further, the high-resolution TEM (HRTEM) analysis reveals the crystallinity of N,S-GQDs. The lattice spacing of 0.23 nm agrees with that of in-plane lattice spacing of graphene (100 facet), which is similar to that of the reported N,S-GQDs [16]. The XRD spectra of LTO/N,S-GQDs and LTO-c were shown in Fig.3. Two kinds of LTO-based materials exhibit a similar XRD pattern. The each XRD spectrum includes seven diffraction peaks at 18.4°, 35.6°, 43.4°, 47.4°, 57.2°, 62.8° and 66.1°, corresponding to crystal planes of LTO (111), LTO (311), LTO (400), LTO (331), LTO (511), LTO (440) and LTO(531). All diffraction peaks in Fig.3 conform to spinel-type LTO (PDF card no. 49-0207) and no characteristic peak of other impurities is observed, suggesting the formation of well crystallized LTO phase. These diffraction peaks are considerably sharp, suggesting highly crystalline

nature of LTO for the two samples. The single crystalline feature will help to increase the electronic conductivity, thus leading to high electrochemical performance for lithium ion lithium. In addition, the peaks at (111) are taken to evaluate the crystallite size of LTO/N,S-GQDs and LTO-c. The crystallite sizes of LTO/N,S-GQDs and LTO-c are calculated via the Scherrer’s equation: (1) where d is the crystallite size,  is the wavelength of x-ray, K is 0.89 as the shape factor,  is the diffraction angle of the peak, and  is the true half-peak width. According to the calculated results, LTO/N,S-GQDs and LTO-c possess the crystalline size of 48.36 and 48.32 nm, respectively. The above results also demonstrates that the addition of small amounts of N,S-GQDs don’t effect on the crystal structure of LTO. Due to low degree of crystallization, N,S-GQDs only offer a weak and wide diffraction peak at

26. The

existence of LTO

decreases the

crystallization degree of N,S-GQDs in the LTO/N,S-GQDs, leading to further reduce the diffraction peak intensity at 26. Therefore, we are difficult to observe an obvious diffraction peak at 26° on the XRD patterns of LTO/N,S-GQDs.

The stoichiometric composition ratios of LTO/N,S-GQDs was clarified by ICP-AES. The chemical composition ration of Li/Ti is found to be 0.802, which is close to the theoretical molar ratio of 0.8. In addition, the yield percentages was calculated by element Ti. The result shows that the yield percentage of LTO/N,S-GQDs is about 99.6. Fig. 4 presents typically XPS patterns of the LTO/N,S-GQDs. There are six peaks at 55.2, 172.2, 284.6, 399.9, 458.9 and 530.5 eV on the total XPS spectrum, indicating that the hybrid is composed of lithium (Li), sulfur (S), carbon (C), nitrogen (N), titanium (Ti) and oxygen (O). Because all C element in the hybrid come from the N,S-GQD used in the synthesis, the weight ratio of C can be used to calculate the content of N,S-GQD in the hybrid. The content of N,S-GQD was found to be 9.2%. There are three peaks on the high-resolution spectrum of N1s XPS spectrum. The peak at 398.4eV could be assigned to N2, corresponds to pyridinic N. The peak at 399.8 eV could be assigned to amide, amine or pyrrolic N. The peak at 401 eV could be assigned to N4 and corresponds to graphitic N. It was noteworthy that the peak intensity of graphitic nitrogen

was

weaker

than

that

of

pyridinic

nitrogen

and

pyrrolic/pyridinic nitrogen, implying that pyridinic nitrogen and pyrrolic nitrogen were dominant in the N,S-GQDs. There are three peaks

on the high-resolution spectrum of C1s XPS spectrum. The binding energy peak at 284.5 eV confirms graphitic structure (sp2 C-C) of the N,S-GQDs. The peak at 285.5 eV suggests the presence of C-O, C-S and C-N, and the peak around 288.0 eV could be assigned to C=O. There are two main peaks on the high-resolution spectrum of Ti2p XPS spectrum. The peaks at about 458 eV and 464 eV could be assigned to Ti 2p3/2 and Ti2p1/2 core level binding energies of Ti4+, respectively. Taking a further look, the spectrum curve of LTO/N,S-GQDs can be fitted into small part of Ti3+ peaks at 456.8 eV and 461.2 eV corresponding to Ti3+2p3/2 and Ti3+2p1/2. Although the peak areas are only 3.2% and 2.8%, respectively, the small amount of Ti3+ can effectively catalyse the redox reaction of Ti 3+/Ti4+ and enhance the bulk conductivity. There are two main peaks on the high-resolution spectrum of S 2p XPS spectrum. Their existence confirms the presence of C-S-C units. Raman spectroscopy of the LTO/N,S-GQDs was presented in Fig.s3. Raman spectrum of carbon materials is characterized by two main features: G band arising from the first order scattering of the E1g phonon of sp2 carbon atoms and D band arising from a breathing mode of point photons of A1g symmetry [25]. For single LTO, no any Raman peaks appear in the range of 1200-3000cm-1 . Thus, the existence of LTO can

not interfere with Raman measurements of the N,S-GQDs in the LTO/N,S-GQDs. Fig.s3 shows that two typical Raman peaks of the N,S-GQDs, including D band (1385 cm-1) and G band (1575cm-1). The fact verifies the existences of N,S-GQDs in the LTO/N,S-GQDs again. In addition, a relative intensity of the “disorder” D-band and the crystalline G-band (ID /IG) value was currently used for evaluating the dispersibility of N,S-GQDs. This is because the peak intensity of D band increases and the intensity of G band decreases with decrease of the crystallization degree of carbon materials. Based on the data in Fig.6, we can calculate the ID/IG value of the N,S-GQD in the hybrid. The ID/IG is 1.42, indicating that the N,S-GQD sheets have been fully dispersed in the hybrid. 3.3. Electrochemical property The electrochemical properties of LTO/N,S-GQDs and LTO-c electrodes were studied by cyclic voltammograms (CV) and electrochemical impedance spectrum (EIS), respectively. The shapes of redox peaks observed in a CV curve can reflect the electrochemical reaction kinetics of the lithium ion intercalation and deintercalation process. A sharp and well-resolved

peak

signifies

fast

lithium

ion

intercalation

and

deintercalation, whereas a broad peak suggests a sluggish process. Fig.s4 shows that the LTO-c electrode gives a weak and wide oxidation and

reduction peak, indicating a relatively sluggish lithium ion intercalation and deintercalation process. However, the LTO/N,S-GQDs electrode provides a higher oxidation and reduction peak current than the LTO-c electrode, indicating an improved lithium ion intercalation and deintercalation process. The enhanced CV performance could be attributed that the introduction of N,S-GQDs into LTO crystal greatly improves the electronic conductivity, which results in the increase in peak current of the LTO/N,S-GQDs electrode. In addition, Fig.s4 also shows that two CV curves give similar oxidation and reduction peak potential, verifying that the addition of N,S-GQDs does not change the electrochemical process of LTO. AC impedance spectra of the LTO-c and LTO/N,S-GQDs electrodes were measured to investigate the kinetic processes. As shown in Fig.5, the recorded Nyquist spectra of LTO-c and LTO/N,S-GQDs cells comprise of one semicircle in high frequencies and a subsequent inclined line at the low frequency end. Here, we employed a simple equivalent circuit embedded in Fig.5 to model the sole semicircle behaviors. In the circuits, Rs is the bulk electrolyte resistance. Rct is the interfacial capacitance. Ci is the interfacial capacitance and Zw is Warburg resistance. The fitting data were listed in Table s1. Table s1 shows that Rs and Rct of the

LTO/N,S-GQDs electrode (4.5 Ω, 3.8 Ω) are much lower than that of the LTO-c electrode (170.7Ω, 234Ω). The result demonstrates that the introduction

of

N,S-GQDs

can

greatly

improve

the

electronic

conductivity, which is consistent with results of the CV investigation. To understand the effect of N,S-GQDs on the electrochemical property, fluorescence decays of the LTO/N,S-GQDs was measured by fluorescence lifetime imaging microscopy. Fig.6 presents typically fluorescence decay curve of the LTO/N,S-GQDs. The experimental data is fitted with a biexponential

decay

function.

The

biexponential

decay

with

approximately equal amplitudes and with lifetimes of τ 1=2.2958 ns and τ2=0.4634 ns is observed. The decay lifetime of τ1 could be assigned to radiative recombination of electrons and holes trapped on the N,S-GQDs [16]. The decay lifetime of τ2 should be attributed to the electronic coupling of N,S-GQDs to LTO. A short lifetime reveals the existence of an ultrafast electron transfer process between the N,S-GQDs and the LTO crystal. Here, it must be pointed out that the doped nitrogen and sulfur in GQDs play important roles in improvement of the electrochemical performance. On the one hand, the nitrogen-doping introduces a new kind of the surface states. Electrons trapped by the new formed surface states are able to facilitate a high yield of radiative recombination [26].

The introduced sulfur atoms would enhance the effect of nitrogen atoms on the properties of the GQDs through a cooperative effect. Such a cooperative effect will significantly accelerate the electron transfer by strong electronic coupling between N,S-GQDs and LTO. As the N,S-GQDs in the hybrid are closely integrated with LTO, the combination with high quality creates rich of the channels for the electron-hole recombination, further improving the electron transfer. On the other hand, the co-doping nitrogen and sulfur are to increase the numbers of sites that lithium ions combine with the electrode materials. Due to the excellent flatness, 2D graphene sheets of GQDs tread to agglomerate into big aggregates by the strong van der waals force between sheets, which will largely reduce the specific surface of GQDs. Such a flat 2D graphene sheet is also easy to closely attach on the surface of LTO crystals. This will further reduce number of the sites that lithium ions combine with the electrode materials and leads to decrease the electrical capacity. However, the co-doping nitrogen and sulfur can well resolve the above problem, because the introduction of nitrogen and sulfur atoms destroy the flatness of graphene sheets. This not only reduces the agglomeration of graphene sheets, but also makes the graphene sheets and LTO crystal have a large number of the electrolyte channels, leading to an ultrafast

mass transport of the electrolyte. Moreover, the co-doping nitrogen and sulfur is also to improve the affinity of LTO electrode materials with the electrolyte. Because of a relatively big electronegativity of nitrogen and sulfur, the doping nitrogen and sulfur can further increase the polarity of GQDs. A more polar electrode material will easier to fully contract with the polar electrolyte and leads to a better the dynamic behavior of the charge/discharge. 3.4. Battery performance Fig. 7 presents the charge/discharge curves of LTO/N,S-GQDs electrode at different rates from 0.1 C to 10 C. The discharge capacity of LTO/N,S-GQDs electrode is 254.2 mAh g -1 at 0.1C. The value is higher than that of LTO/GQDs electrode (225.7 mAh g-1) and LTO-c electrode (169.9 mAh g-1), indicating that the introduction of N,S-GQDs will greatly increase the specific capacity. The previous investigations have demonstrated that graphene as an electrode material of lithium ion battery can produce the capacity of more than 1000 mAh g -1, which is much higher than the theoretical capacity of pure phase LTO [27]. Compared to common graphene, N,S-GQDs have a bigger specific surface area and more carbon atoms on the edge of its small graphene sheets. This will bring a higher storage lithium capacity. Thus, the addition of

N,S-GQDs results in an increase of the specific capacity. In addition, we also observe from Fig.7 only at 0.1C the discharge capacity is obviously more than the charge capacity, and at other rates the discharge capacity is very close to the charge capacity with a good coulombic efficiency. The discharge curve at 0.1C is the first discharge curve of the half-cell, the enhanced discharge capacity should be attributed to the formation of SEI film

during

the

first

discharge

process.

This

is

because

the

electron-active N,S-GQDs may react with the electrolyte to form the SEM film, which will result in an enhanced discharge capacity. As two control samples, the charge/discharge curves of LTO-c electrode and LTO/GQDs electrode were measured at different rates from 0.1C to 10C, respectively. The LTO-c electrode shows a flat operation potential plateau when charged/discharge at low rates such as 0.1 C and 0.5 C. The potential plateau of LTO-c electrode becomes shorter and gradually bends down with the rate increased, while that of the LTO/N,S-GQDs electrode still remains flat. This is because the polarization of LTO-c would be increased at an increased rate. A comparison of potential difference (E) between the charge and discharge plateau potentials was taken for the LTO-c electrode, LTO/GQDs electrode and LTO/N,S-GQDs electrode in the study. The results were listed in Table s2. The E value

represents the degree of polarization of the electrode. A bigger E value means a bigger polarization and poor reaction kinetics, while a small E value means a lower polarization and better reaction kinetics. From Table s2, we can see that the E for the LTO/N,S-GQDs electrode is far less than that of LTO-c electrode, and slightly less than that of LTO/GQDs electrode at all discharge rates from 0.1 C to 10 C, indicating a lower polarization and better reaction kinetic. The improvement could mainly be attributed that the introduction of N,S-GQDs. As a new type of QDs, the GQDs offer an unique property associated with both graphene and QDs. This will greatly enhance the electronic conductivity, leading to improve the polarization of the electrode. In addition, the co-doping nitrogen and sulfur further into graphene sheets improve the electrochemical performance of GQDs, which will further improve polarization of the electrode. Fig. 8 presents the discharge specific capacity with number of cycles for the LTO/N,S-GQDs, LTO-G and LTO-c at different rates from 0.1 C to 10 C. Three electrodes display good cyclic performance at the each rate, which could be attributed to stable cycle life of LTO. Compared with the LTO-c electrode, the LTO/N,S-GQDs electrode exhibits a higher specific capacity and better high-rate capability as shown in Fig.8. The capacity

difference between the LTO-C and LTO/N,S-GQDs can be clearly seen and becomes larger at larger rates. For instance, at a 1 C rate, the discharge specific capacity of LTO/N,S-GQDs is 186.2 mAh g-1, which is about 1.69-fold that of LTO-c. While at 10 C, the discharge specific capacity of LTO/N,S-GQDs is 126.5 mAh g -1, which is close to 2.86-fold that of LTO-c. The improved high-rate capability for LTO/N,S-GQDs electrode could be explained by the reduced resistance and polarization of the electrode, as described above. With N,S-GQDs conductive sheets throughout the whole hybrid material including the crystal interior and exterior, the rich of pathways for the electron transfer are produced, thus the electron transfer is more effective and electronic conductivity of the electrode is greatly improved. In addition, after the low current rate of 1C is used again, the specific capacities of three cells are well restored, which suggests LTO/N,S-GQDs, LTO-G and LTO-c exhibit good cycle performance and electrochemical stability. Interestingly, both the specific discharge capacity and high-rate performance of LTO/N,S-GQDs are obviously better than that of the LTO-G electrode. This could be mainly attributed to difference in sheet size between common graphene and N,G-GQDs. On the hand, the graphene is composed of micron scale sheets. The use of graphene sheets may seriously hinder the electrolyte

transport due to its relatively big size and results in the reduce of the specific discharge capacity, especially at high rate. However, the N,S-GQDs have the minimal size of graphene sheets. Its effect on the electrolyte transport can be neglected in all rate. On the other hand, the co-doping nitrogen and sulfur in the GQD sheets brings a strong electronic coupling between N,S-GQDs and LTO, which will significantly accelerate the electron transfer. Moreover, the N,S-GQDs sheets in the hybrid are well dispersed on the surface and interior of LTO crystal, which are closely integrated with LTO. The combination with high quality creates rich of the channels for the electron-hole recombination in the hybrid, further enhancing the electron transfer [28]. For the above reasons, the LTO/N,S-GQDs electrode can offer largely enhanced capacity and high-rate performance. Cyclic performance of LTO/N,S-GQDs at 2C was investigated and the result was listed in Fig.9. A stable cycle life can be observed from this curve. In the first cycle, the discharge capacity was 186.2 mAh g-1, and after 2000 charge/discharge cycles it still remains at 180.5 mAh g-1 with the coulombic efficiency of 99.6-99.9%, a 96.9% retention of the first discharge capacity, indicating excellent cycle performance.

Conclusions In the study, we have successfully fabricated a new lithium titanate/nitrogen and sulfur co-doped graphene quantum dots. The hybrid offers fully crystalline interconnected porous framework composed of nanoscale LTO crystals. The unique architecture achieve to maximize the high-rate performance and enhance the power density. The introduction of N,S-GQDs greatly accelerates the electron/charge transfer, owing to co-doping nitrogen and sulfur and their minimal size. The

LTO/N,S-GQDs

anode

exhibits

excellent

electrochemical

performance for lithium ion battery and can be widely used as the anode material for high-performance lithium ion batteries. The study also provides an attractive approach for building on the LTO-based electrode materials for various energy storage devices. Acknowledgements The authors acknowledge the financial support from Prospective Joint Research Project: Cooperative Innovation Fund (No.BY2014023-01), the country “12th Five-Year Plan” to support science and technology project (No. 2012BAK08B01), National Natural Science Foundation of China

(No.21176101), Fundamental Research Funds for Central Universities (No.JUSRP51314B) and MOE & SAFEA for the 111 Project (B13025).

References [1] S. Tan, Y.J. Ji, Z.R. Zhang, Y. Yang, Recent progress in research on high-voltage electrolytes for lithium-ion batteries, Chemphyschem 15 (2014) 1956-1969. [2] C.R. Zhu, D.L. Chao, J. Sun, I.M. Bacho, Z.X. Fan, C.F. Ng, X.H. Xia, H. Huang, H. Zhang, Z.X. Shen, G.Q. Ding and H.J. Fan, Enhanced Lithium Storage Performance of CuO Nanowires by Coating of Graphene Quantum

Dots,

Adv.

Mater.

Interfaces

2

(2015)

DOI: 10.1002/admi.201400499. [3] X.C. Sun, P.V. Radovanovic, B. Cui, Advances in spinel Li4Ti5O12 anode materials for lithium-ion batteries, New J. Chem. 39 (2015) 38-63. [4] T.F. Yi, S.Y. Yang, Y. Xie, Recent advances of Li4Ti5O12 as a promising next generation anode material for high power lithium-ion batteries, J. Mater. Chem. A 3 (2015) 5750–5777.

[5] J.M. Kim, H.T. Chung, Role of transition metals in layered Li[Ni, Co, Mn]O2 under electrochemical operation, Electrochim. Acta 49 (2004) 3573-3580. [6] S.W. Han, J.Y. Jeong, D.H. Yoon, Effects of high-energy milling on the solid-state synthesis of pure nano-sized Li4Ti5 O12 for high power lithium battery applications, Appl. Phys. A 114 (2014) 925-930. [7] J.M. Feckl, K. Fominykh, M. Doblinger, D. Fattakhova-Rohlfing, T. Bein, Nanoscale porous framework of lithium titanate for ultrafast lithium insertion, Angew. Chem. Int. Edit. 51 (2012) 7459-7463. [8] Q.Y. Zhang, M.G. Verde, J.K. Seo, X. Li, Y. S. Meng, Structural and electrochemical properties of Gd-doped Li4Ti5O12 as anode material with improved rate capability for lithium-ion batteries, J. Power Sources 280 (2015) 355-362. [9] Z.X. Liu, L.M. Sun, W.Y. Yang, J.B. Yang, S.B. Han, D.F. Chen, Y.T. Liu, X.F. Liu, The synergic effects of Na and K co-doping on the crystal structure and electrochemical properties of Li4 Ti5O12 as anode material for lithium ion battery, Solid State Sciences 44 (2015) 39-44. [10] S.Y. Yang, J. Yuan, Y.R. Zhu, T.F. Yi, Y. Xie, Structure and electrochemical properties of Sc3+-doped Li4Ti5O12 as anode materials for lithium-ion battery, Ceramics International 41 (2015) 7073-7079.

[11] W. Li, M.Z. Chen, J.J. Jiang, R. Wu, F. Wang, W.J. Liu, G.C. Peng and M.Z. Qu, Structural and electrochemical characteristics of SiO2 modified Li4Ti5O12 as anode for lithium-ion batteries, J. Alloys Compd. 637 (2015) 476-482 [12] M. Krajewski, M. Michalska, B. Hamankiewicz, D. Ziolkowska, K.P. Korona, J.B. Jasinski, M. Kaminska, L. Lipinska, A. Czerwinski, Li4Ti5O12 modified with Ag nanoparticles as an advanced anode material in lithium-ion batteries, J. Power Sources 245 (2014) 764-771. [13] J. Liu, K.P. Song, P.A. Van Aken, J. Maier, Y. Yu, Self-Supported Li4Ti5O12-C nanotube arrays as high-rate and long-life anode materials for flexible Li-ion batteries, Nano Lett. 14 (2014) 2597-2603. [14] Y. Oh, S. Nam, S. Wi, J. Kang, T. Hwang, S. Lee, H.H. Park, J. Cabana, C. Kim, B. Park, Effective wrapping of graphene on individual Li4Ti5O12 grains for high-rate Li-ion batteries, J. Mater. Chem. A 2 (2014) 2023-2027. [15] C.C. Chen, Y. Huang, H. Zhang, X.F. Wang, G.Y. Li, Y.J. Wang, L.F. Jiao, H.T. Yuan, Small amount of reduce graphene oxide modified Li4Ti5O12 nanoparticles for ultrafast high-power lithium ion battery, J. Power Sources 278 (2015) 693-702. [16] Y.Q. Dong, H.C. Pang, H.B. Yang, C.X. Guo, J.W. Shao, Y.W. Chi, C.M. Li, T. Yu, Carbon-based dots Co-doped with nitrogen and sulfur for high quantum

yield and excitation-independent emission, Angew. Chem.-Int. Edit. 52 (2013) 7800-7804. [17] P. Roy, A.P. Periasamy, C. Chuang, Y.R. Liou, Y.F. Chen, J. Joly, C.T. Liang, H.T. Chang, Plant leaf-derived graphene quantum dots and applications for white LEDs, New Journal of Chemistry 38 (2014) 4946-4951. [18] W.W. Liu, Y.Q. Feng, X.B. Yan, J.T. Chen, Q.J. Xue, Superior Micro-Supercapacitors Based on Graphene Quantum Dots, Adv. Funct. Mater. 23 (2013) 4111-4122. [19] X. Zhou, Z. Tian, J. Li, H. Ruan, Y. Ma, Z. Yang, Y. Qu, Synergistically enhanced activity of graphene quantum dot/multi-walled carbon nanotube composites as metal-free catalysts for oxygen reduction reaction, Nanoscale 6 (2014) 2603-2607. [20] Z.C. Ding, Z. Hao, B. Meng, Z.Y. Xie, J. Liu and L.M. Dai, Few-layered graphene quantum dots as efficient hole-extraction layer for high-performance polymer solar cells, Nano Energy 15 (2015) 186-192. [21] Z.S. Qian, X.Y. Shan, L.J. Chai, J.R. Chen, H. Feng, A fluorescent nanosensor based on graphene quantum dots-aptamer probe and graphene oxide platform for detection of lead (II) ion, Biosens. Bioelectron. 68 (2015) 225-231. [22] I.A. Luk’yanchikova, D.V. Ivlev, M.G. Kiselev, G.A. Al’per, Concentration

dependence of the viscosity of tert-butanol-water mixtures: Physical experiment and computer simulation, Russ. J. Gen. Chem+ 74 (2004) 1256-1162. [23] T. Yan, R.Y. Li, Z.J. Li, J.K. Liu, G.L. Wang, Z.Q. Gu, A free template strategy for the fabrication of nickel/cobalt double hydroxide microspheres with tunable nanostructure and morphology for high performance supercapacitors, RSC Adv. 3 (2013) 19416-19422. [25] A.C. Ferrari, J. Robertson, Interpretation of Raman spectra of disordered and amorphous carbon, Phys. Rev. B 61 (2000)14095-14099. [26] K. Dokko, M. Mohamedi, M. Umeda, I. Uchida, Kinetic study of Li-ion extraction and insertion at LiMn2O4 single particle electrodes using potential step and impedance methods, J. Electrochem. Soc. 150 (2003) A425-A429. [27] J. Liang, Y. Jiao, M. Jaroniec, S. Z. Qiao, Sulfur and nitrogen dual-doped mesoporous graphene electrocatalyst for reduction with synergistically enhanced performance, Angewandte Chemie 124 (2012) 11664-11668. [28] V.S. Channu, R. Bobba, R. Holze, Graphite and graphene oxide electrodes for lithium ion batteries, Colloid Surf. A-Physicochem. Eng. Asp. 436 (2013) 245-251. [29] D.N. Tafen, O.V. Prezhdo, Size and temperature dependence of electron transfer between CdSe quantum dots and a TiO2 Nanobelt, J. Phys. Chem. C

119 (2015) 5639-5647.

Figure captions Fig.1 Procedure for the synthesis of LTO/N,S-GQDs Fig.2 SEM (a and b) and TEM images (c and d) of the LTO/N,S-GQDs and TEM (e) and HRTEM (f) of N,S-GQDs in the LTO/N,S-GQDs Fig.3 XRD patterns of LTO-c and LTO/N,S-GQDs Fig.4 Total, N1s, Ti2p, C1s and S2p XPS spectra of LTO/N,S-GQDs Fig.5 AC impedance spectra of the LTO-c (a) and LTO/N,S-GQDs (b) with the equivalent circuit from the EIS measurements (inset) Fig.6 The fluorescent decays of LTO/N,S-GQDs Fig.7 The charge/discharge curves of LTO/N,S-GQDs electrode at 0.1C/0.1C (a), 0.2C/0.2C (b), 0.5C/0.5C (c), 1 C/1C (d), 1.5C/1.5C (e), 2C/2C (f), 5C/5C (g) and 10C/10C (i) Fig.8 The specific discharge capacities of LTO/N,S-GQDs (a), LTO-G (b) and LTO-c (c) electrodes at different rates Fig.9 The cyclic performance of LTO/N,S-GQDs at 1 C

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Fig. 7

Fig. 8

Fig. 9

Graphical abstract