Improvement of Electrochemical Performance of Li[Ni 0.8 Co 0.15 Al 0.05 ]O 2 Cathode

Improvement of Electrochemical Performance of Li[Ni 0.8 Co 0.15 Al 0.05 ]O 2 Cathode
Improvement of Electrochemical Performance of Li[Ni 0.8 Co 0.15 Al 0.05 ]O 2 Cathode

MATERIALS AND INTERFACES

Improvement of Electrochemical Performance of Li[Ni 0.8Co 0.15Al 0.05]O 2Cathode Materials by AlF 3coating at Various Temperatures

Byung-Chun Park,?Hyung-Bae Kim,?Hyun Joo Bang,?Jai Prakash,?and Yang-Kook Sun*,?

Department of Chemical Engineering,Center for Information and Communication Materials,Hanyang Uni V ersity,Seoul 133-791,South Korea,and Department of Chemical and Biological Engineering,Illinois Institute of Technology,10West 33rd Street,Chicago,Illinois 60616

A thin AlF 3layer of ~10nm was uniformly coated on the particle surface of Li[Ni 0.8Co 0.15Al 0.05]O 2.The AlF 3coating improved cycle performance at 55,30,and –10°C and improved storage characteristics at 60°C.In studies of differential capacity (d Q/d V )versus voltage,the AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2showed little variation in redox peaks with cycling.Electrochemical impedance spectroscopy suggested that the AlF 3coating played an important role in stabilizing the interface between the cathode and the electrolyte.

Introduction

The high energy density and power capability of Li-ion batteries have made them a popular power source for portable electronic devices and,more recently,hybrid electric vehicles (HEVs)and plug-in hybrid electric vehicles (PHEVs).1,2Ni-rich Li[Ni 1-x M x ]O 2(M )metal)cathode materials are alterna-tives to LiCoO 2and have lower cost,lower toxicity,and higher reversible capacity.3–5However,Ni-rich Li[Ni 1-x M x ]O 2has been found to exhibit low thermal stability,poor cycling behavior,and increased impedance during cycling,particularly at high temperatures.6,7One of the most promising materials is Li[Ni 0.8Co 0.15Al 0.05]O 2.8,9The substitution of Ni with Co and Al improves its structural and thermal stability without sacri?c-ing discharge capacity.In spite of the improved electrochemical performance of the Li[Ni 0.8Co 0.15Al 0.05]O 2,the major barriers that hinder commercial-scale application are its insuf?cient cycle life and safety concerns under abuse conditions,especially at temperatures above 55°C.There have been several approaches to solve these problems.One method is to completely encap-sulate the Ni-based active materials with Li[Ni 0.5Mn 0.5]O 2,a stable active material.10,11Another problem inherent to Li[Ni 0.8Co 0.15Al 0.05]O 2with high Ni content is its rapid reaction to moisture and ambient CO 2,resulting in the formation of LiOH and Li 2CO 3on the cathode particle surface.12,13Li extraction from the lattice gives rise to cation ion mixing,in turn leading to severe capacity fading and reduced reversible capacity.One way to improve the electrochemical performance is to modify the cathode particle surface with metal oxide 14–18and metal ?uoride 19–21to reduce the catalytic activity of the transition metal with a high oxidation state in LiMO 2(M )Co,Ni,Mn).The electrochemical properties and thermal stability of the coated cathode materials show signi?cant improvement over uncoated ones.Though the reason is not well-understood,the protection of the cathode surface provided by the coating may be responsible for the enhanced electrochemical

performance.One plausible reason for the improvement is that the coating layer protects against HF attack of the electrolyte,suppressing the decomposition of the active material in cases where pristine particles would be severely degraded by cycling.17,18,22The preservation of the cathode surface stabilizes the host structure of the cathode materials,allowing for stable charge-transfer resistance during successive charge/discharge cycles.19–21

In this paper,we study the effect of an AlF 3coating over Li[Ni 0.8Co 0.15Al 0.05]O 2on electrochemical performance at vari-ous temperatures.We also investigate the reason for the improvement of the electrochemical performance in the presence of the coating by comparing the differential capacity (d Q /d V)versus voltage pro?les of pristine and AlF 3-coated Li[Ni 0.8-Co 0.1Mn 0.1]O 2.Experimental Section

[Ni 0.8Co 0.15Al 0.05](OH)2powders were prepared by copre-cipitation,as reported previously.23An aqueous solution of NiSO 4·6H 2O,CoSO 4·7H 2O,and Al 2(SO 4)3·16H 2O was pumped into a continuously stirred tank reactor (CSTR,capacity 4L)in an N 2atmosphere.Simultaneously,an NaOH solution (aq)and an appropriate amount of NH 4OH solution (aq),a chelating agent,were fed separately into the reactor.The spherical [Ni 0.8Co 0.15Al 0.05](OH)2powders were dried at 110°C for 24h to remove adsorbed water.Finally,a mixture of the dehydrated [Ni 0.8Co 0.15Al 0.05](OH)2and LiOH ·H 2O were preheated to 480°C for 5h and then heated at 750°C for 20h in an oxygen ?ow.The chemical compositions of the prepared powders were analyzed using atomic absorption spectroscopy (AAS,Vario 6,Analyticjena).To coat the surface of the Li[Ni 0.8Co 0.15Al 0.05]O 2with AlF 3,ammonium ?uoride (Aldrich)and aluminum nitrate nonahydrate (Aldrich)were separately dissolved in distilled water.After the prepared Li[Ni 0.8Co 0.15Al 0.05]O 2powders were immersed in the aluminum nitrate nonahydrate solution,the ammonium ?uoride solution was slowly added.The molar ratio of F to Al was ?xed at 6,and the amount of AlF 3in the solution corresponded to 0.25mol %of the Li[Ni 0.8Co 0.15Al 0.05]O 2powders.The solution containing the active material was

*Corresponding author.Phone:+82-2-2220-0524.Fax:+82-2-2282-7329.E-mail:yksun@hanyang.ac.kr.?

Hanyang University.?

Illinois Institute of Technology.

Ind.Eng.Chem.Res.2008,47,3876–3882

387610.1021/ie0715308CCC:$40.75 2008American Chemical Society

Published on Web 04/29/2008

continuously stirred for 5h at 80°C,accompanied by a slow evaporation of solvent.The Li[Ni 0.8Co 0.15Al 0.05]O 2powders obtained were heated at 400°C for 5h in an oxygen ?ow.Powder X-ray diffraction (XRD)(Rigaku,Rint-2000)employing Cu K R radiation was used to identify the crystalline phase of the prepared powders at each stage.The surface of the AlF 3-coated powder was also observed using transmission electron microscopy (TEM,JEOL 2010).Differential scanning calorim-etry (DSC)experiments were carried out for the positive electrode materials by fully charging the coin cell to 4.3V at a constant current and voltage and opening it in an Ar-?lled dry room.The measurements were carried out in a differential scanning calorimeter 200PC (NETZSCH,Germany)using a temperature scan rate of 1°C/min.Charge–discharge tests were performed with a coin-type cell (CR2032)having a current density between 40and 100mA/g at 30°C.The cell consisted of positive and negative electrodes of lithium metal separated by a porous polypropylene ?lm.To fabricate the cathode,the prepared powders were mixed with carbon black and polyvi-nylidene ?uoride (85:7.5:7.5)in N -methylpyrrolidinon.The slurry obtained was coated onto Al foil and roll-pressed at 120°C in air.The electrodes were dried at 120°C overnight in a vacuum state prior to use.Preliminary cell tests were conducted using a 2032coin-type cell using Li metal for the anode.The electrolyte solution used was 1M LiPF 6in ethylene carbonate -diethyl carbonate (1:1in volume,Cheil Industries,Inc.,Korea).Results and Discussion

Figure 1shows the XRD patterns of the pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2materials.It was con?rmed that the two powders have a well-de?ned layer structure based on the hexagonal a -NaFeO 2structure with space group Rm .The lattice constants,a and c,of the AlF 3-coated Li[Ni 0.8Co 0.15-Al 0.05]O 2powders were 2.865(5)and 14.194(7)?,as calculated by a least-squares method.These values approximate those of the pristine material,a )2.864(1)and c )14.191(0)?,indicating that the AlF 3was not incorporated into the host structure.Chemical analysis showed that the prepared powder composition was Li[Ni 0.796Co 0.153Al 0.051]O 2.

Figure 2shows a TEM image of the AlF 3-coated Li[Ni 0.8-Co 0.15Al 0.05]O 2powder.The TEM image proves the presence of the uniform AlF 3coating layer of 10nm on the particle surface.Energy-dispersive X-ray spectroscopy con?rmed the composition of the coating layer to be rich in Al.

Figure 3displays the initial charge/discharge curves of the Li/pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2cells by applying a constant current of 40mA/g (0.5C-rate)between 3.0and 4.3V vs Li +/Li.The charge voltage of the AlF 3-coated

Li[Ni 0.8Co 0.15Al 0.05]O 2electrode shows lower resistance than the pristine Li[Ni 0.8Co 0.15Al 0.05]O 2in all operating regions,which increases the discharge capacity and the rate capability.The initial discharge capacity of the AlF 3-coated Li[Ni 0.8Co 0.15-Al 0.05]O 2is 178mAh/g,as compared to that of the pristine material,which has a capacity of 168mAh/g.It is speculated that a thin AlF 3-coating layer lowers the charge-transfer resistance,which facilitates Li +intercalation/deintercalation at the interface between the cathode and the electrolyte.19–21

Parts a and b of Figure 4show the discharge capacity of Li/pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2cells over 50cycles at 30and 55°C,respectively.The cells were ?rst activated by applying a current density of 0.2C-rate (20mA/g)for two cycles and were then cycled at 0.5C-rate (40mA/g)between 3.0and 4.3V vs Li +/Li.During the 30°C cycling,the AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2electrode delivered a discharge capacity of 173mAh/g,while the pristine electrode delivered only 160mAh/g.The pristine Li[Ni 0.8Co 0.15Al 0.05]O 2showed a gradual decrease in capacity,leading to a capacity retention of 91%after 50cycles.The AlF 3-coated Li[Ni 0.8-Co 0.15Al 0.05]O 2displayed a slightly improved capacity retention of 97%after the same cycling period.However,the cycling behavior of the AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2at 55°C was remarkably enhanced,showing capacity retention of 94%after 62cycles.The pristine electrode showed a gradual decrease in capacity,leading to a capacity retention of only 73%.Recently,Abraham et al.reported that a Li x Ni 1-x O-type layer was observed on the surface of a Li[Ni 0.8Co 0.2]O 2particle on which surface layers were 2-5nm thick in samples from 0%power fade cells,and >35nm thick in samples from 43%power fade cells.6It is believed that the thin AlF 3coating layer lowers the catalytic activity of Ni 4+with electrolyte species in highly delithiated Li 1-x [Ni 0.8Co 0.15Al 0.05]O 2and,thus,reduces the formation of

the

Figure 1.XRD patterns of (a)pristine and (b)AlF 3-coated Li[Ni 0.8Co 0.15-Al 0.05]O 2

.

Figure 2.Bright-?eld TEM images of (a)pristine and (b)AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2

.

Figure 3.Initial charge–discharge curves of Li/pristine and Li/AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2cells at a current density of 40mA/g between 3.0and 4.3V.

Ind.Eng.Chem.Res.,Vol.47,No.11,20083877

Li x Ni 1-x O-type layer resulting from Ni 4+decomposition,espe-cially at temperatures higher than 55°C.

Figure 5shows a plot of discharge capacity versus cycle number for the Li/pristine and AlF 3-coated Li[Ni 0.8Co 0.15-Al 0.05]O 2between 3.0and 4.3V at 0.2C-rate (20mA/g)and –10°C.The cells were activated for two cycles at 0.2C-rate and 30°C before the test.Although AlF 3-coated Li[Ni 0.8Co 0.15-Al 0.05]O 2showed a slightly larger discharge capacity (123mAh/g)than the pristine material (113mAh/g),the two electrodes showed a similar capacity retention of 67–69%at –10°C as compared to 30°C.However,the cycling stability was remark-

ably enhanced in the electrode with the AlF 3coating,which showed no capacity loss,while the pristine electrode suffered from a capacity loss of 14%after only 20cycles.It is believed that the improved performance at low temperatures can be attributed to the lower charge-transfer resistance.The data are consistent with our previous reports showing that Li/AlF 3-coated LiCoO 2,Li[Ni 1/3Co 1/3Mn 1/3]O 2,and Li[Ni 0.8Co 0.1Mn 0.1]O 2cells exhibit a signi?cantly lower charge-transfer resistance (R ct )than pristine electrodes during extended cycling.19–21

The storage characteristics of cathode materials at an elevated temperature,especially in the delithiated state,are of great interest in judging their practical use for lithium secondary batteries.Figure 6shows the variations in discharge capacities of Li/pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2cells in the charged state of 4.3V at 60°C as a function of storage time.The prepared cells were preliminarily charged and discharged between 3.0and 4.3V at 0.2C-rate and 30°C for two cycles to determine their initial capacities.The cells were then charged to 4.3V at 60°C and stored for 24,36,and 72h.The aged cells were cooled down to 30°C,and their discharge capacity was measured at each storage interval.While the pristine Li[Ni 0.8Co 0.15Al 0.05]O 2showed a rapid decrease in capacity,leading to a capacity retention of only 46%after 72h in storage,the storage capacity of the AlF 3-coated electrode greatly improved,showing a capacity retention of 91%after the same storage time.The data also show the impact of AlF 3coating.To investigate the effects of transition metal dissolution of the Li 1-δ[Ni 0.8Co 0.15Al 0.05]O 2on storage time,the chemical compositions with storage times were analyzed by AAS,and the results are shown in Table 1.The Ni and Co dissolution from the pristine Li 0.2[Ni 0.8Co 0.15Al 0.05]O 2was increased with storage time,while dissolution from the AlF 3-coated Li 0.2[Ni 0.8Co 0.15Al 0.05]O 2was kept nearly constant.For

example,

Figure 4.Variation of the discharge capacities of Li/pristine and Li/AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2cells at (a)30and (b)55°C between 3.0and 4.3V at 0.5C-rate (40

mA/g).

Figure 5.Plot of discharge capacity vs cycle number for the Li/pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2between 3.0and 4.3V at 0.2C-rate (20mA/g)and –10°

C.

Figure 6.Variation of the discharge capacities of (a)Li/pristine and (b)AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2cells in the charged state of 4.3V at 60°C as a function of storage time.

3878Ind.Eng.Chem.Res.,Vol.47,No.11,2008

the chemical compositions of the pristine and AlF 3-coated Li 0.2[Ni 0.8Co 0.15Al 0.05]O 2after 72h were estimated to be Li 0.42[Ni 0.70Co 0.103Al 0.04]O 2-δand Li 0.27[Ni 0.75Co 0.14Al 0.04]O 2-δ,respectively.Note that the Li content of the pristine materials was quite a bit larger than that of the AlF 3-coated materials.Recently,Ozawa et al.reported that capacity fading in the charged state originates from (i)the chemical reactions involving electron transfer between the electrolyte and the electrode and (ii)Li-intercalation from the electrolyte into the cathode host structure.24It is believed that the AlF 3coating is so resistant to HF attack on the cathode particle surface that dissolution of the transition metal is greatly reduced,which in turn enhances the storage performance at 60°C.

Figure 7shows the discharge capacities of the Li/pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2cells as a function of various C-rates (40-1000mA/g)between 3.0and 4.3V.The capacity retention percentage of each C-rate (compared with 0.2C-rate)is also shown in Figure 7.The cells were charged galvanostati-cally with a current density of 20mA/g (0.2C-rate)before each discharge test and then discharged at current densities from 20(0.2C-rate)to 1000mA/g (5C-rate).As observed in Figure 7,the capacity retention difference between the pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2electrode became larger with increasing C-rate.For example,the capacity retentions of AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2were 92%and 62%at 1and 5C-rate,respectively,while those of the pristine material were 89%and 54%.Notice that the AlF 3-coated Li[Ni 0.8Co 0.15-Al 0.05]O 2electrode delivers a larger discharge capacity than the pristine one across the range of C-rates.This result leads us to believe that the AlF 3-coating layer on Li[Ni 0.8Co 0.15Al 0.05]O 2plays an important role in expediting the Li +intercalation to the host structure by stabilizing the structure and reducing the interfacial resistance between the cathode and the electrolyte.19–21

Figure 8shows differential capacity (d Q /d V)versus voltage pro?les of Li/pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2cells at the ?rst,30th,and 50th cycles.The cells were cycled between 3.0and 4.3V at a current density of 100mA/g.During the ?rst cycle,the Li/pristine Li[Ni 0.8Co 0.15Al 0.05]O 2cell showed a sharp oxidation peak at 3.68V and a satellite peak at 3.76V,

Table 1.Chemical Compositions of Pristine and AlF 3-coated Li 1-δ[Ni 0.8Co 0.15Al 0.05]O 2Charged to 4.3V with Storage Time

samples

36

72

pristine Li 0.24[Ni 0.8Co 0.15Al 0.05]O 2

Li 0.34[Ni 0.72Co 0.13Al 0.05]O 2-δLi 0.42[Ni 0.70Co 0.10Al 0.04]O 2-δAlF 3-coated Li 0.24[Ni

0.8Co 0.15Al 0.05]O 2

Li 0.25[Ni 0.76Co 0.15Al 0.05]O 2-δ

Li 0.27[Ni 0.75Co 0.14

Al 0.04]O 2-δ

Figure 7.Rate capability test of Li/pristine and Li/AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2cells at various C-rates.

Figure 8.Differential capacity vs voltage

of (a)Li/pristine and (b)Li/AlF 3-coated Li[Ni 0.8Co 0.1Mn 0.1]O 2cells in the voltage range of 3.0–4.3V and 30°C.

Figure 9.Differential capacity vs voltage of (a)Li/pristine and (b)Li/AlF 3-coated Li[Ni 0.8Co 0.1Mn 0.1]O 2cells in the voltage range of 3.0–4.3V and 60°C.

Ind.Eng.Chem.Res.,Vol.47,No.11,20083879

as well as several other peaks during charging due to multiphase transitions.25,26Corresponding redox peaks at 3.67,3.93,and 4.13V were also observed.The redox peaks became more polarized and shifted farther apart.For example,the oxidation peak height at 3.68V and the corresponding reduction peak at 3.67V decreased after 50cycles,indicating surface degradation of the host structure and/or increase of the interfacial impedance between the cathode and the electrolyte.20,21No peak variation was observed for the AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2elec-trode during cycling.

Figure 9shows the discharge differential capacity (d Q/d V )versus voltage pro?les of Li/pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2cells charged to 4.3V at 60°C as a function of storage time.Both the pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2electrodes showed three distinct reduction peaks at 3.68,3.94,and 4.16V at 30°C,which corresponded to a phase transition between two O 3structures.25However,the two electrodes showed different reduction peak changes after storage.For the pristine Li[Ni 0.8Co 0.15Al 0.05]O 2electrode,storage at 60°C caused increased polarization of the reduction peaks and shifted them farther apart.For example,the peaks at 3.68and 3.94V shifted to 3.31and 3.3V,respectively,after a 72h storage time.Notice that the reduction peak at 4.16V corre-sponding to the H2-H3phase change disappeared after 36h of storage.Surprisingly,the reduction peaks for the AlF 3-coated

Li[Ni 0.8Co 0.15Al 0.05]O 2remained almost unchanged even after a 72h storage time,though the peak positions moved to a lower angle.

To study the reason for the improved electrochemical performances of the AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2,elec-trochemical impedance spectroscopy (EIS)was carried out for the pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2in a charged state of 4.3V as a function of cycle number at 55°C and storage time at 60°C.Figure 10shows Nyquist plots of the pristine and coated materials.Expanded views of the two overlapped semicircles in the high-to-medium frequency region are also shown in the inset of Figure 10.The equivalent circuit used in this study was reported in a previous study.The scattering symbol represents experimental data,and the continuous lines represent ?tted results using the equivalent circuit.These simulated lines correspond well to the plots of observed data.The semicircle occurring at a high frequency could be attributed to the resistance of the surface ?lm (R sf ),and the second circle appearing at high-to-medium frequency is associated with the charge-transfer resistance (R ct ).Notice that the charge-transfer resistance (Li[Ni 0.8Co 0.15Al 0.05]O 2/electrolyte interface)for both electrodes rapidly increased with cycling,while the surface-?lm resistance was kept almost stable at values of 30–40?(Table 1).Similar phenomena were observed for AlF 3-coated LiCoO 2,Li[Ni 1/3Co 1/3Mn 1/3]O 2,and Li[Ni 0.8Co 0.1Mn 0.1]O 2,as we previously reported.19–21Variations with cycle number in surface-?lm resistance (R sf )and charge-transfer resistance (R ct )for both the pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2electrodes are shown in Table 2.The R ct value of the pristine Li[Ni 0.8Co 0.1Mn 0.1]O 2electrode after the ?rst cycle was 15.2?and increased rapidly to 167.0?after the 50th cycle.However,the R ct of the AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2electrode held nearly constant,growing from 14.9?after the ?rst cycle to 89.0?after the 50th cycle.

Figure 11represents the Nyquist plots of the pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2with storage time at

60

Figure 10.Nyquist plots of (a)the Li/pristine and (b)the Li/AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2cell at the 1st,25th,and 50th cycles.

Table 2.Surface-Film Resistance (R sf )and Charge-Transfer

Resistance (R ct )for Pristine and AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2with Cycle Number (Units Are ?)

pristine

Li[Ni 0.8Co 0.15Al 0.05]O 2

AlF 3-coated

Li[Ni 0.8Co 0.15Al 0.05]O 2

cycle number

R sf R ct R sf R ct 129.115.238.914.92531.7106.128.931.350

33.5

167.0

31.3

89.0

Figure 11.Nyquist plots of (a)the Li/pristine and (b)the Li/AlF 3-coated Li[Ni 0.8Co 0.15Al 0.05]O 2cell at fresh,36h,and 72h of storage time.

3880Ind.Eng.Chem.Res.,Vol.47,No.11,2008

°C in the charged state of4.3V.Similar impedance growth was observed for the coated electrode after storage time. However,the R sf values of both electrodes slowly increased with storage time,unlike the electrodes with cycling shown in Figure10.These results indicate that the Li[Ni0.8Co0.15Al0.05]O2has poor storage characteristics at elevated temperatures.Variations in R sf and R ct with storage time for both the pristine and AlF3-coated Li[Ni0.8Co0.15Al0.05]O2electrodes are given in Table3.The R ct value of the pristine Li[Ni0.8Co0.15Al0.05]O2electrode after the?rst cycle was38.6?and increased rapidly to95.3?after the50th cycle.However,the R ct of AlF3-coated Li[Ni0.8-Co0.15Al0.05]O2held nearly constant,growing from11.0?after the?rst cycle to23.0?after the50th cycle.Also,the R sf of AlF3-coated Li[Ni0.8Co0.15Al0.05]O2had a lower value than that of pristine Li[Ni0.8Co0.15Al0.05]O2.The enhanced storage performance of AlF3-coated Li[Ni0.8Co0.15Al0.05]O2 at60°C could be attributed to reduced charge-transfer and surface-?lm resistances.It is clear from these results that AlF3coating gives rise to a signi?cant lowering of charge-transfer resistance during cycling and storage time at elevated temperatures,which in turn enhances the cycling behavior and the rate capability.

Conclusions

Li[Ni0.8Co0.15Al0.05]O2particles were uniformly coated with a thin AlF3layer.The AlF3-coated Li[Ni0.8Co0.15Al0.05]O2 electrode showed improved electrochemical properties after storage at60°C,improved cycling performance at25,55, and–10°C,and improved rate capability.The AlF3-coated Li[Ni0.8Co0.15Al0.05]O2showed a capacity retention of91% after72h of storage,even at60°C,while the capacity retention of the pristine material decreased to only46%.From the differential capacity(d Q/d V)versus voltage pro?les,we see that an AlF3coating on cathode particles does not suppress structural changes induced by phase transition but stabilizes the host structure.Such excellent electrochemical properties could originate from the low and stable charge-transfer resistance between cathode and electrolyte and the suppression of surface degradation of the Li[Ni0.8Co0.15Al0.05]O2host structure by the AlF3coating. Acknowledgment

This work was supported by KOSEF through the Research Center for Energy Conversion and Storage.

Supporting Information Available:Included in the Sup-porting Information is a graph of pulse power ASI as a function of DID for a Li/pristine and AlF3-coated Li[Ni0.8Co0.15Al0.05]O2 cell.This material is available free of charge via the Internet at https://www.360docs.net/doc/af118675.html,.

Literature Cited

(1)Amine,K.;Liu,J.;Kang,S;Belharouak,I.;Hyung,Y.;Vissers,

D.;Henriksen,G.Improved lithium manganese oxide spinel/graphite Li-ion cells for high-power applications.J.Power Sources.2004,129,14–19.

(2)Andersson,A.M.;Abraham,D.P.;Haasch,R.;MacLaren,S.;Liu, J.;Amine,K.Surface Characterization of Electrodes from High Power Lithium-Ion Batteries.J.Electrochem.Soc.2002,149,A1358–A1369.

(3)Ohzuku,T.;Ueda,A.;Nagayama,M.Electrochemistry and Structural Chemistry of LiNiO2(R3m)for4V Secondary Lithium Cells.J. Electrochem.Soc.1993,140,1862–1870.

(4)Dahn,J.R.;Fuller,E.W.;Obrovac,M.;von Sacken,U.Thermal stability of Li x CoO2,Li x NiO2andλ-MnO2and consequences for the safety of Li-ion.Solid State Ionics1994,69,265–270.

(5)Arai,H.;Okada,S.;Sakurai,Y.;Yamaki,J.-I.Thermal behavior of Li1-y NiO2and the decomposition mechanism.Solid State Ionics1998,109, 295–302.

(6)Abraham,D.P.;Twesten,R.D.;Balasubramanian,M.;Petrov,I.; McBreen,J.;Amine,K.Surface changes on LiNi0:8Co0:2O2particles during testing of high-power lithium-ion https://www.360docs.net/doc/af118675.html,mun.2002,4, 620–625.

(7)Arai,H.;Tsuda,M.;Saito,K.;Hayashi,M.;Sakurai,Y.Thermal Reactions Between Delithiated Lithium Nickelate and Electrolyte Solutions. J.Electrochem.Soc.2002,146,A401–A406.

(8)Guilmard,M.;Pouillerie,C.;Croguennec,L.;Delmas,C.Structural and electrochemical properties of LiNi0.70Co0.15Al0.15O2.Solid State Ionics 2003,160,39–50.

(9)Weaving,J.S.;Coowar,F.;Teagel,D.A.;Cullen,J.;Dass,D.A.; Bindin,P.;Green,R.;Macklin,W.J.Development of high energy density Li-ion batteries based on LiNi1-x-y Co x Al y O2.J.Power Sources2001,97 (98),733–735.

(10)Sun,Y.-K.;Myung,S.-T.;Kim,M.-H.;Prakash,J.;Amine,K. Synthesis and Characterization of Li[(Ni0.8Co0.1Mn0.1)0.8(Ni0.5Mn0.5)0.2]O2 with the Micro-scale Core-shell Structure as the Positive Electrode Material for Lithium Batteries.J.Am.Chem.Soc.2005,127(38),13411–13418.

(11)Sun,Y.-K.;Myung,S.-T.;Park,B.-C.;Amine,K.Synthesis of Spherical Nano to Micro-Scale Core-Shell Particle Li[(Ni0.8Co0.1Mn0.1)1-x(Ni0.5Mn0.5)x]O2and Its Application to Lithium Batteries.Chem.Mater. 2006,18(22),5159–5163.

(12)Zhuang,G.V.;Chen,G.;Shim,J.;Song,X.;Ross,P.N.; Richardson,T.J.Li2CO3in LiNi0.8Co0.15Al0.05O2cathodes and its effects on capacity and power.J.Power Sources2004,134,293–297.

(13)Matsumoto,K.;Kuzuo,R.;Takeya,K.;Yamanaka,A.Effects of CO2in air on Li deintercalation from LiNi1-x-y Co x Al y O2.J.Power Sources 1999,1–82,558–561.

(14)Kweon,H.-J.;Kim,G.-B.;Park,D.-G.K.R.Patent Appl.,1998/ 0012005.

(15)Cho,J.;Kim,T.-J.;Kim,Y.J.;Park,B.Zero-Strain Intercalation Cathode for Rechargeable Li-Ion Cell.Angew.Chem.Int.Ed.2001,40, 3367–3369.

(16)Miyashiro,H.;Yamanaka,A.;Tabuchi,M.;Seki,S.;Nakayama, M.;Ohno,Y.;Kobayashi,Y.;Mita,Y.;Usami, A.;Wakihara,M. Improvement of Degradation at Elevated Temperature and at High State-of-Charge Storage by ZrO2Coating on LiCoO2.J.Electrochem.Soc.2006, 153,A348–A353.

(17)Sun,Y.-K.;Lee,Y.-S.;Yoshio,M.;Amine,K.Synthesis and Electrochemical Properties of ZnO-coated LiNi0.5Mn1.5O4Spinel Materials for Lithium Secondary Batteries.Electrochem.Solid-State Lett.2002,5, A99–A102.

(18)Sun,Y.-K.;Hong,K.-J.;Prakash,J.The Effect of ZnO Coating on Electrochemical Cycling Behavior of Spinel LiMn2O4Cathode Materials at Elevated Temperature.J.Electrochem.Soc.2003,150,A970–A972.

(19)Sun,Y.-K.;Han,J.-M.;Myung,S.-T.;Lee,S.-W.;Amine,K. Signi?cant Improvement of High Voltage Cycling Behavior AlF3-Coated https://www.360docs.net/doc/af118675.html,mun.2006,8,821–826.

(20)Sun,Y.-K.;Cho,S.-W.;Lee,S.-W.;Yoon,C.S.;Amine,K. Improvement of High Voltage Cycling Performance of AlF3-coated Li[Ni1/ 3Co1/3Mn1/3]O2Cathode Materials for Lithium Secondary Batteries.J. Electrochem.Soc.2007,154,A168–A172.

(21)Woo,S.-U.;Yoon,C.S.;Amine,K.;Belharouak,I.;Sun,Y.-K. Signi?cant Improvement of Electrochemical Performance of AlF3-coated Li[Ni0.8Co0.1Mn0.1]O2Cathode Materials.J.Electrochem.Soc.2007,154 (11),A1005–A1009.

(22)Myung,T.;Izumi,K.;Komaba,S.;Sun,Y.-K.;Yashiro,H.; Kumagai,N.Role of Alumina Coating on Li-Ni-Co-Mn-O Particles as Positive Electrode Material for Lithium-Ion Batteries.Chem.Mater.2005, 17,3695–3704.

Table3.Surface-Film Resistance(R sf)and Charge-Transfer Resistance(R ct)for Pristine and AlF3-coated Li[Ni0.8Co0.15Al0.05]O2 with Storage Time(Units Are?)

pristine

Li[Ni0.8Co0.15Al0.05]O2

AlF3-coated

Li[Ni0.8Co0.15Al0.05]O2

storage time R sf R ct R sf R ct

fresh12.438.613.511.0

36h16.768.115.013.6

72h25.095.318.623.0

Ind.Eng.Chem.Res.,Vol.47,No.11,20083881

(23)Lee,M.-H.;Kang,Y.-J.;Myung,S.-T.;Sun,Y.-K.Synthetic Optimization of Li[Ni1/3Co1/3Mn1/3]O2via Co-precipitation.Electrochim. Acta2004,50,939–948.

(24)Gabrisch,H.;Ozawa,Y.;Yazami,R.Crystal structure studies of thermally aged LiCoO2and LiMn2O4cathodes.Electrochim.Acta2006, 52,1499–1506.

(25)(W)Li,J.N.Reimers and J.R.Dahn.In situ x-ray diffraction and electrochemical studies of Li1-x NiO2.Solid State lonics1993,67,123–130.

(26)Chen,Z.;Dahn,J.R.Methods to obtain excellent capacity retention in LiCoO2cycled to4.5V.Electrochim.Acta2004,49,1079–1090.

Recei V ed for re V iew November9,2007

Re V ised manuscript recei V ed February21,2008

Accepted March1,2008

IE0715308

3882Ind.Eng.Chem.Res.,Vol.47,No.11,2008

钳型电流表的使用方法及注意事项

https://www.360docs.net/doc/af118675.html, 钳型电流表的使用方法及注意事项 一、使用说明 平常一些我们了利用普通电流表测量电流时,需要将电路切断停机后才能将电流表接入进行测量,这是很麻烦的,有时正常运行的电动机是不允许这样做的。此时,使用钳形电流表就显得方便多了,可以在不切断电路的情况下来测量电流。 钳形电流表工作原理: 钳形电流表是由电流互感器和电流表组合而成。电流互感器的铁心在捏紧扳手时可以张开;被测电流所通过的导线可以不必切断就可穿过铁心张开的缺口,当放开扳手后铁心闭合。穿过铁心的被测电路导线就成为电流互感器的一次线圈,其中通过电流便在二次线圈中感应出电流。从而使二次线圈相连接的电流表便有指示-----测出被测线路的电流。钳形

https://www.360docs.net/doc/af118675.html, 表可以通过转换开关的拨档,改换不同的量程。但拨档时不允许带电进行操作。钳形表一般准确度不高,通常为2.5——5级。为了使用方便,表内还有不同量程的转换开关供测不同等级电流以及测量电压的功能。 二、使用方法 1、在使用钳形电流表前应仔细阅读说明书,弄清是交流还是交直流两用钳形表。 2、钳形表每次只能测量一相导线的电流,被测导线应置于钳形窗口中央,不可以将多相导线都夹入窗口测量。 3、被测电路电压不能超过钳形表上所标明的数值,否则容易造成接地事故,或者引起触电危险。 4、使用高压钳形表时应注意钳形电流表的电压等级,严禁用低压钳形表测量高电压回路的电流。用高压钳形表测量时,应由两人操作,非值班人员测量还应填写第二种工作票,测量时应戴绝缘手套,站在绝缘垫上,不得触及其它设备,以防止短路或接地。 5、在高压回路上测量时,禁止用导线从钳形电流表另接表计测量。测量高压电缆各相电流时,电缆头线间距离应在300mm以上,且绝缘良好,待认为测量方便时,方能进行。 6、观测表计时,要特别注意保持头部与带电部分的安全距离,人体任何部分与带电体的距离不得小于钳形表的整个长度。 三、注意事项 (1)被测线路的电压要低于钳表的额定电压。 (2)当电缆有一相接地时,严禁测量。防止出现因电缆头的绝缘水平低发生对地击穿爆炸而危及人身安全。

实验四 linux-2.6.35内核的编译和配置

实验四 linux-2.6.35内核的编译和配置 【实验目的】 了解内核的编译过程及配置选项的内容 【实验环境】 1、 Ubuntu 10.10发行版 2、 u-boot-2010.03 3、 FS2410平台 4、 交叉编译器 arm-none-linux-gnueabi-gcc-4.3.2 【实验步骤】 实验步骤中的1-4,已经做过就不要重复了 1、 将实验代码中的rootfs.tar.bz2解压到/source 下,已经做过就不要重复了 $ tar xvf rootfs.tar.bz2 –C /source 2、 解压内核并进入内核目录 $ tar xvf linux-2.6.35.tar.bz2 $ cd linux-2.6.35 3、 修改Makefile 修改linux-2.6.35 目录下的Makefile ,找到 ARCH ?= $(SUBARCH) CROSS_COMPILE ?= 改为 ARCH ?= arm CROSS_COMPILE ?= arm-none-linux-gnueabi- 4、 配置内核 设置平台 设置编译工具

make menuconfig Kernel Features ---> [*] Use the ARM EABI to compile the kernel [*] Allow old ABI binaries to run with this kernel (EXPERIMENTAL) 5、添加驱动文件 将实验代码2410GPIO_TEST_26/2410GPIO_TEST_drv.c拷贝到drivers/char下 6、修改对应Kconfig 修改drivers/char/Kconfig,在menu "Character devices"下面 加入如下内容: config 2410GPIO_TEST_DRV tristate "S3C2410 test drv Device Support" depends on ARCH_S3C2410 ---help--- support led test device driver on FS2410 develop board 7、修改对应Makefile 在drivers/char/Makefile 中 找到在obj-$(CONFIG_HANGCHECK_TIMER) += hangcheck-timer.o , 在其下一行添加: obj-$(CONFIG_2410GPIO_TEST_DRV) += 2410GPIO_TEST_drv.o 8、静态编译内核 ?配置内核时按“空格”选择,配置完成后保存退出 $ make menuconfig Device Drivers ---> Character devices ---> <*> S3C2410 test drv Device Support ?重新编译内核并把内核拷贝到tftpboot下 $ make zImage

内核参数的修改方法

Linux内核参数修改方法 由于Linux的内核参数信息都存在内存中,因此可以通过命令直接修改,并且修改后直接生效。但是,当系统重新启动后,原来设置的参数值就会丢失,而系统每次启动时都会自动去/etc/sysctl.conf文件中读取内核参数,因此将内核的参数配置写入这个文件中,是一个比较好的选择。 首先打开/etc/sysctl.conf文件,查看如下两行的设置值,这里是: kernel.shmall = 2097152 kernel.shmmax = 4294967295 如果系统默认的配置比这里给出的值大,就不要修改原有配置。同时在/etc/sysctl.conf文件最后,添加以下内容: fs.file-max = 6553600 kernel.shmmni = 4096 kernel.sem = 250 32000 100 128 net.ipv4.ip_local_port_range = 1024 65000 net.core.rmem_default = 4194304 net.core.rmem_max = 4194304 net.core.wmem_default = 262144 net.core.wmem_max = 262144 这里的“fs.file-max = 6553600”其实是由“fs.file-max = 512 * PROCESSES”得到的,我们指定PROCESSES的值为12800,即为“fs.file-max =512 *12800”。 sysctl.conf文件修改完毕后,接着执行“sysctl -p”使设置生效。 [root@localhost ~]# sysctl -p 常用的内核参数的含义如下。 kernel.shmmax:表示单个共享内存段的最大值,以字节为单位,此值一般为物理内存的一半,不过大一点也没关系,这里设定的为4GB,即 “4294967295/1024/1024/1024=4G”。 kernel.shmmni:表示单个共享内存段的最小值,一般为4kB,即4096bit. kernel.shmall:表示可用共享内存的总量,单位是页,在32位系统上一页等于4kB,也就是4096字节。 fs.file-max:表示文件句柄的最大数量。文件句柄表示在Linux系统中可以打开的文件数量。 ip_local_port_range:表示端口的范围,为指定的内容。 kernel.sem:表示设置的信号量,这4个参数内容大小固定。 net.core.rmem_default:表示接收套接字缓冲区大小的缺省值(以字节为单位)。 net.core.rmem_max :表示接收套接字缓冲区大小的最大值(以字节为单位) net.core.wmem_default:表示发送套接字缓冲区大小的缺省值(以字节为单位)。 net.core.wmem_max:表示发送套接字缓冲区大小的最大值(以字节为单位)。

必须懂的53个电脑英文缩写

·PC:个人计算机Personal Computer ·CPU:中央处理器Central Processing Unit ·CPU Fan:中央处理器的“散热器”(Fan) ·MB:主机板MotherBoard ·RAM:内存Random Access Memory,以PC-代号划分规格,如PC-133,PC-1066,PC-2700 ·HDD:硬盘Hard Disk Drive ·FDD:软盘Floopy Disk Drive ·CD-ROM:光驱Compact Disk Read Only Memory ·DVD-ROM:DVD光驱Digital Versatile Disk Read Only Memory ·CD-RW:刻录机Compact Disk ReWriter ·VGA:显示卡(显示卡正式用语应为Display Card) ·AUD:声卡(声卡正式用语应为Sound Card) ·LAN:网卡(网卡正式用语应为Network Card) ·MODM:数据卡或调制解调器Modem ·HUB:集线器

·WebCam:网络摄影机 ·Capture:影音采集卡 ·Case:机箱 ·Power:电源 ·Moniter:屏幕,CRT为显像管屏幕,LCD为液晶屏幕 ·USB:通用串行总线Universal Serial Bus,用来连接外围装置·IEEE1394:新的高速序列总线规格Institute of Electrical and Electronic Engineers ·Mouse:鼠标,常见接口规格为PS/2与USB ·KB:键盘,常见接口规格为PS/2与USB ·Speaker:喇叭 ·Printer:打印机 ·Scanner:扫描仪 ·UPS:不断电系统 ·IDE:指IDE接口规格Integrated Device Electronics,IDE接口装置泛指采用IDE接口的各种设备

钳形表与三相四线表使用方法及注意事项

一、功率与电能测量方法: 1.1 功率测量方法 (1). 直接法:测量功率可直接用电动系功率表、数字功率表或三相功率表,测量三相功率还可以用单相功率表接成两表法或三表法,虽然有求和过程,但一般仍将它归为直接法. (2). 间接法:直流可通过测量电压、电流间接求得功率。交流则需要通过电压、电流和功率因数求得功率。 1.2 电能测量方法 (1). 直接法:直接测量电能,直流可使用电动系电能表,交流用感应系或电子电能表。 (2). 间接法:电能测量一般不用间接法,只有在功率稳定不变的情况下用功率表和记时时钟进行测量。 二.钳形电流表的应用 钳形电流表按结构原理不同分为磁电式和电磁式两种,磁电式可测量交流电流和交流电压;电磁式可测量交流电流和直流电流。钳形表俯视图如图1.1所示: 图1.1 钳形表俯视图

钳形表机械图如图1.2所示: 图1.2 钳形表机械视图 2.1.钳形电流表的使用方法和使用时应注意的事项 (1). 在进行测量时用手捏紧扳手即张开,被测载流导线的位置应放在钳口中间,防止产生测量误差,然后放开扳手,使铁心闭合,表头就有指示。 (2). 测量时应先估计被测电流或电压的大小,选择合适的量程或先选用较大的量程测量,然后再视被测电流、电压大小减小量程,使读数超过刻度的1/2,以便得到较准确的读数。 (3). 为使读数准确,钳口两个面应保证很好的接合,如有杂声,可将钳口重新开合一次,如果声音依然存在,可检查在接合面上是否有污垢存在,如有污垢,可用汽油擦干净。 (4). 测量低压可熔保险器或低压母线电流时,测量前应将邻近各相用绝缘板隔离,以防钳口张开时可能引起相间短路。 (5). 有些型号的钳形电流表附有交流电压刻度,测量电流、电压时应分别进行,不能同时测量。 (6). 不能用于高压带电测量。 (7). 测量完毕后一定要把调节开关放在最大电流量程位置,以免下次使用时由于未经选择量程而造成仪表损坏。

VMware 虚拟机存储管理

VMware 虚拟机存储管理 1)实现虚拟机共享存储 VMware vSphere环境中对共享存储的访问是通过VMware vStorage VMFS 实现的,这是一种专为虚拟机设计的高性能集群文件系统。 VMware vStorage VMFS 是专为虚拟服务器环境而设计、构造和优化的,可让多个虚拟机对由集群式存储构成的整合池进行共享访问,从而提高资源利用率。VMware vStorage VMFS 还为分布式基础架构服务奠定了基础,例如虚拟机和虚拟磁盘文件实时迁移,以及分布式资源调度、整合备份和自动灾难恢复。 作为文件系统,VMware vStorage VMFS 将构成虚拟机的所有文件存储在一个目录中。经过优化,可以支持大型文件,同时也可以执行许多小型的并发写操作。通过自动处理虚拟机文件,VMware vStorage VMFS 对整个虚拟机进行封装,使其很容易成为灾难恢复解决方案的一部分。事实上,VMware Infrastructure 3 之所以被TechTarget 评为“2006 年度灾难恢复产品”,VMware vStorage VMFS 是主要原因之一。 作为逻辑卷管理器,VMware vStorage VMFS 实现了一个存储资源界面,使得多种类型的存储(SAN、iSCSI 和NAS)能够以可承载虚拟机的数据存储的形式出现。通过以聚合存储资源方式实现那些数据存储的动态增长,VMware vStorage VMFS 可提供在最少停机或无停机的情况下增加共享存储资源池的能力。 VMware vStorage VMFS 与传统文件系统 传统文件系统在指定时间只允许一台服务器对同一文件进行读写访问。与之相对,VMware vStorage VMFS 使用共享存储来允许多个VMware ESX 实例对同一存储资源进行并发读写访问。 VMware vStorage VMFS 利用分布式日志来允许跨这些多服务器资源池进行快速、弹性的恢复。此外,VMware vStorage VMFS 提供了进行灾难恢复所必需的虚拟机快照功能,并且是VMware Consolidated Backup (VCB) 用来提供虚拟环境代理备份的界面。 VMware vStorage VMFS 与CFS 和CVM VMware vStorage VMFS 并不包含当今的其他集群文件系统(CFM) 和集群卷管理(CVM)

操作系统实验1(编译内核)实验指导书

实验一虚拟机平台下的Linux内核编译 步骤一、实验准备:将windows下的Linux-2.4.32内核复制到虚拟机中。 一般有三种方法可以实现:虚拟磁盘、文件共享、网络设置。 下面介绍的是虚拟磁盘的方法: 我们已经将Linux-2.4.32内核源代码的压缩文件放入虚拟磁盘,并在FTP上共享。 1.请从FTP服务器上将实验指导书以及虚拟磁盘拷贝到E盘根目录下。 2. windows平台下启动虚拟机VMware, VMWare中安装虚拟硬盘: 1)安装前建议先把内存调大一些,这样后面编译的速度会快一些。(Memory,调到512M) 2)点击Add 3)点击Next

上述步骤完成后点击Finish 。 观察发现:虚拟机中将出现两个磁盘,一个是原有的磁盘,另一个是虚拟磁盘,相当于对本机外接了一个磁盘。(请将设置完成后的虚拟机中出现两个磁盘的界面截图,一个是6G ,另一个106M )。

4.启动Linux操作系统 在启动过程中可以见到当前只有一个可选择内核,内核版本是Linux2.4.20。我们本次实验的目的是在虚拟机下重新编译一个新的内核,内核版本是Linux2.4.32.完成本实验后,重启虚拟机将出现两个可选择内核。 5.将虚拟磁盘安装到/mnt/mydisk目录下.(mount命令),如图: 1)先用fdisk –l命令查看虚拟磁盘是否“连接”成功。记下虚拟磁盘名称(/dev/sdb)。 2)进入到mnt目录下创建mydisk目录,该目录作为访问虚拟磁盘的入口,此时该目录为空。 3)返回根目录root,并安装虚拟磁盘:mount /dev/sdb /mnt/mydisk

linux 内核参数修改

linux 内核参数修改 配置 Linux 内核参数(2种方法),修改后不用重启动更新: /sbin/sysctl -p 第一种:打开/etc/sysctl.conf 复制如下内容 kernel.shmall = 2097152 kernel.shmmax = 2147483648 kernel.shmmni = 4096 kernel.sem = 250 32000 100 128 fs.file-max = 65536 net.ipv4.ip_local_port_range = 1024 65000 net.core.rmem_default=262144 net.core.wmem_default=262144 net.core.rmem_max=262144 net.core.wmem_max=262144 第二种:打开终端 cat >> /etc/sysctl.conf< kernel.shmall = 2097152 kernel.shmmax = 2147483648 kernel.shmmni = 4096 kernel.sem = 250 32000 100 128 fs.file-max = 65536 net.ipv4.ip_local_port_range = 1024 65000 net.core.rmem_default=262144 net.core.wmem_default=262144 net.core.rmem_max=262144 net.core.wmem_max=262144 EOF 这里,对每个参数值做个简要的解释和说明。 (1)shmmax:该参数定义了共享内存段的最大尺寸(以字节为单位)。缺省为32M,对于oracle来说,该缺省值太低了,通常将其设置为2G。(2)shmmni:这个内核参数用于设置系统范围内共享内存段的最大数量。该参数的默认值是 4096 。通常不需要更改。 (3)shmall:该参数表示系统一次可以使用的共享内存总量(以页为单位)。缺省值就是2097152,通常不需要修改。(共享内存段的数量,以页为主,每个页是4K) (4)sem:该参数表示设置的信号量。一般大于maxproc的一点就行了。 (5)file-max:该参数表示文件句柄的最大数量。文件句柄设置表示在linux系统中可以打开的文件数量。 修改好内核以后,执行下面的命令使新的配置生效。 [root @linux1 /root]# /sbin/sysctl -p 以 root 用户身份运行以下命令来验证您的设置: /sbin/sysctl -a | grep shm /sbin/sysctl -a | grep sem /sbin/sysctl -a | grep file-max /sbin/sysctl -a | grep ip_local_port_range 例如: # /sbin/sysctl -a | grep shm kernel.shmmni = 4096 kernel.shmall = 2097152 kernel.shmmax = 2147483648

编译内核实验报告

实验一编译Linux内核 实验时间 6小时 实验目的 认识Linux内核的组成,掌握配置、编译、安装Linux内核的步骤。 实验目标 下载2.6.19或更新的Linux内核,配置该内核使其支持NTFS,并在新的内核中修改其版本为Linux NameTestKernel x.x.x,其中,Name是你的名字(汉语拼音);x.x.x是新内核的版本号,最后在你的机器上编译安装这个新内核。 背景知识 参见《Red Hat Enterprise Linux 4入门与提高》第20章。 实验步骤 1.验证gcc的可用:在你自己的工作目录下,编译链接运行Hello World程序。 2.在https://www.360docs.net/doc/af118675.html,上下载指定的内核,或者查找更新的稳定版内核并 下载之。 3.准备相关工具。 提示:如当前运行的Linux内核是基于2.4版本的,则需要更新以下软件: module-init-tools和mkinitrd。具体更新信息可参见下载内核源代码中的 Documentation/Changes这个文件。 4.把源代码解压缩至/usr/src中,最终形成/usr/src/linux x.x.x/目录(x.x.x是新 内核的版本号)。 提示:这里的注意点是路径的选择,一般要放在/usr/src/linux x.x.x/目录下面,以满足Makefile对路径设置的初始要求。 5.进入源代码的根目录,找到合适自己的内核配置方法,并按照实验目标对其 进行配置。

6.修改/usr/src/linux x.x.x/linux/include/linux/verson.h文件中的版本信息。 7.编译内核。 8.安装模块文件。 9.安装内核文件。 10.重新启动新内核。 实验结果 1.实验步骤1中,编译链接运行程序你下达了哪些命令? 2.实验步骤2中,你下载了哪个版本的内核文件? 3.实验步骤3中,你是否安装了相关工具?如安装,则写出安装过程。 4.实验步骤4中,你是用哪些命令解压缩内核文件的? 5.实验步骤5中,你用了哪种内核配置的方法? 6.你对实验步骤6中涉及的文件做了怎样的修改? 7.实验步骤7-9的过程,是否出现错误?如有,你是如何解决的? 8.观察你机器中GRUB的配置文件,它在安装完新内核后发生了哪些变化? 9.新内核启动过程是否成功?如有错误,是哪些错误?你是如何消错的?

内核配置与裁剪

内核配置与裁剪 1. Linux内核配置 内核配置的方法很多,make config、make xconfig、make menuconfig、make oldconfig 等等,它们的功能都是一样的。这里用的是make menuconfig。 过去基于2.x的内核为用户提供了四种基本的内核设置编辑器: ?. config 服务于内核设置的一个冗长的命令行界面; ?. oldconfig 一个文本模式的界面,主要包含一个已有设置文件,对用户所发现的内核资 源中的设置变量进行排序; ?. menuconfig 一个基于光标控制库的终端导向编辑器,可提供文本模式的图形用户界 面; ?. xconfig 一个图形内核设置编辑器,需要安装X-Window系统。 前三种编辑器在设置2.6内核时仍可使用,在运行“make xconfig”后,原有的界面被两个新的图形设置编辑器所代替。这需要具体的图形库和X-Window系统的支持。另外,用户还可以通过“make defconfig”命令,利用所有内核设置变量的缺省值自动建立一个内核设置文件。 下面具体介绍Linux内核配置选项: . 代码成熟度选项 Code maturity level options ---> [*] Prompt for development and/or incomplete code/drivers [*] Select only drivers expected to compile cleanly 在内核中包含了一些不成熟的代码和功能,如果我们想使用这些功能,想打开相关的配置选项,就必需打开这一选项。 . 通用设置选项 General setup ---> () Local version - append to kernel release [*] Automatically append version information to the version string [*] Support for paging of anonymous memory (swap)

为VMwareESXi主机添加本地存储的过程注意事项

为VMware ESXi 主机添加本地存储的过程及注意事项 在规划数据中心的时候, 除了考虑当前的实际情况下, 还要考虑将来二至三年的可能出现的 问题。但有的时候,虽然已经做了多种考虑,但在信息化实施的过程中,仍然会碰到问题, 这时就需要管理员根据实际情况进行处理。 例如,在规划虚拟化数据中心的时候,作为群集中的虚拟化主机, 一般不配置本地硬盘,而 是采用共享存储,以实现系统的高可用性。即时为服务器配置本地硬盘, 一般也只是配置一 个较小的磁盘,例如60?120GB 的固态硬盘安装系统。相信这是大多数虚拟化数据中心的选 择与规划。 对于初期规划没有考虑配置本地存储的服务器,如果在后期需要在服务器安装大容量的存 储,并且需要在现有业务不中断的情况下进行, 应该怎么做?在添加本地存储中碰到问题怎 么解决,本文将讨论这个问题。 1己有数据中心服务器添加本地存储思路及方法 1.1现有数据中心拓扑 近期有一个 vSphere 数据中心,该中心有三台 IBM 3650安装 VMware ESXi 5.5的服务器、1 图1 存储、服务器连接示意图 台IBM V3500存储,在服务器与存储之间使用 SAS 连接,拓扑如图1所示。 rm 3650 M4 172. 16- 3 11M 幽0删 宙IZ2. IG L 16.2 im lTO, Id IG. 203 ESCiOh 172. 16 L& I im: 172. l(i 15 201 172. JR 年为貉700 24 千狂丄屋交搂肌 vSphere Client 轉理E 作站

这三台服务器原来本地没有配置硬盘,是使用的存储分配的空间用于系统及数据。 现在用户的需求是:为每台服务器添加6块硬盘,作为本地的存储。因为这三台ESXi组成一个HA并且已经有虚拟机运行,所以要求在添加本地硬盘的过程中,业务不能中断。另外,由于在开始规划的时候,没有考虑为服务器添加本地硬盘,所以没有选择配置支持RAID5 的RAID卡,故在添加本地硬盘时,需要一同添加RAID卡。 1.2解决思路及原理 在VMware vSphere数据中心中,当所有的虚拟机运行在共享存储上、并且为vSphere数据中心配置了”群集”后,当群集中主机多于2台时,如果要对其中的主机进行维护,包括关机 检修、添加配件等,可以将检修的主机置于"维护模式”,此时该主机上运行的虚拟机会自动"迁移”到其他主机,然后可以重新启动或关闭处于"维护模式”的ESXi主机,在关机后可以 为服务器添加硬件(例如添加RAID卡、添加硬盘、重新配置RAID卡)。在本案例中,就使 用这一功能,可以保证业务不中断的前提下,关闭虚拟化主机(安装了VMware ESXi 5.5 系统),为主机添加RAID卡、本地硬盘,甚至为主机重新安装VMware ESXi 5.5系统并重 新加入现有群集,都不会造成业务的中断。 主要步骤如下: (1)使用vSphere Client (或vSphere Web Client ),登录vCenter Server ,将其中一 台主机,例如第1台主机置于”维护模式”,等该主机上的所有虚拟机,都迁移到其他主机上 之后,关闭该主机。 (2)等主机关闭后,断开该主机电源。拆开机箱,插入RAID卡。 (3)安装好硬盘,重新开机,进入RAID卡配置界面,为添加的硬盘,配置RAID。配置完RAID之后,重新启动主机。 (4)重新进入VMwareESXi系统,将新添加的磁盘,添加到ESXi本地存储。然后将当前主机退出”维护模式”。 (5)之后参照(1)?(4)的步骤,将其他主机进入维护模式,关机,添加RAID卡、硬盘,并配置RAID,重新启动系统并退出维护模式,这些不一一介绍。 (6 )最后检查服务器、存储、及软件系统,没有故障之后,添加存储完成。下面较为详细的介绍整个过程及注意事项。 1.3将主机进入维护模式 使用vSphere Client将要维护的主机进入维护模式,之后关闭主机并断开电源,添加卡,打开 RAID 电源、添加硬盘。主要步骤如下。

Linux 2.6.19.x内核编译配置选项简介(2)

Linux 2.6.19.x内核编译配置选项简介(2) Security Marking 对网络包进行安全标记,类似于nfmark,但主要是为安全目的而设计,如果你不明白的话就别选 Network packet filtering (replaces ipchains) Netfilter可以对数据包进行过滤和修改,可以作为防火墙("packet filter"或"proxy-based")或网关(NAT)或代理(proxy)或网桥使用.选中此选项后必须将"Fast switching"关闭,否则将前功尽弃 Network packet filtering debugging 仅供开发者调试Netfilter使用 Bridged IP/ARP packets filtering 如果你希望使用一个针对桥接的防火墙就打开它 Core Netfilter Configuration 核心Netfilter配置(当包流过Chain时如果match某个规则那么将由该规则的target来处理,否则将由同一个Chain中的下一个规则进行匹配,若不match所有规则那么最终将由该Chain的policy进行处理) Netfilter netlink interface 允许Netfilter在与用户空间通信时使用新的netlink接口.netlink Socket是Linux用户态与内核态交流的主要方法之一,且越来越被重视 Netfilter NFQUEUE over NFNETLINK interface 通过NFNETLINK接口对包进行排队 Netfilter LOG over NFNETLINK interface 通过NFNETLINK接口对包记录.该选项废弃了ipt_ULOG和ebg_ulog机制,并打算在将来废弃基于syslog 的ipt_LOG和ip6t_LOG模块 Layer 3 Independent Connection tracking 独立于第三层的链接跟踪,通过广义化的ip_conntrack支持其它非IP协议的第三层协议 Netfilter Xtables support 如果你打算使用ip_tables,ip6_tables,arp_tables之一就必须选上 "CLASSIFY" target support 允许为包设置优先级,一些排队规则(atm,cbq,dsmark,pfifo_fast,htb,prio)需要使用它 "CONNMARK" target support 类似于"MARK",但影响的是连接标记的值 "DSCP" target support 允许对ip包头部的DSCP(Differentiated Services Codepoint)字段进行修改,该字段常用于Qos "MARK" target support 允许对包进行标记(通常配合ip命令使用),这样就可以改变路由策略或者被其它子系统用来改变其行为"NFQUEUE" target Support 用于替代老旧的QUEUE(iptables内建的target之一),因为NFQUEUE能支持最多65535个队列,而QUEUE 只能支持一个 "NOTRACK" target support 允许规则指定哪些包不进入链接跟踪/NA T子系统 "SECMARK" target support

Linux设置内核参数的方法

Linux设置内核参数的方法 1内核参数的查看方法 使用“sysctl -a”命令可以查看所有正在使用的内核参数。内核参数比较多(一般多达500项),按照前缀主要分为以下几大类:net.ipv4、net.ipv6、net.core、vm、fs、dev.parport、dev.cdrom 、dev.raid、kernel等等。相同的linux,安装的组件和使用的方式不一样,正在使用的内核参数是不一样的。 所有的内核参数的说明文档是放到/usr/src/linux/Documentation/sysctl中的,如果想知道对内核参数的说明,可以到该目录下查看相应的说明文档。 2内核参数的的设置方法 由于Linux的内核参数信息都存在内存中,因此可以通过命令直接修改,并且修改后直接生效。也可以通过文件的方式进行设置。下面就介绍这两种修改方法。 2.1命令设置的方式 可以用两种方法实现。 1、使用“sysctl -w 参数名=值”的方式 假设我们把net.ipv4.ip_forward的值修改为1,使用命令“sysctl -w net.ipv4.ip_forward=1”。 2、修改内核参数对应的proc文件 内核参数位于/proc/sys/之下,参数名称是以文件所在的路径,并将“/”以“.”来取代。举例来说,/proc/sys/net/ip_forward的参数名称为net.ipv4.ip_forward。 同样把net.ipv4.ip_forward的值修改为1,使用命令“echo “1”> /proc/sys/net/ipv4/ip_forward”。 注意,这里proc文件跟普通的文件不一样。一般一个文件用echo写入内容之后,会变成一个文本文件,但echo修改proc文件之后还是个空文件。 2.2文件设置的方式 更改的内核参数默认保存在/etc/sysctl.conf文件中。修改的时候可以直接用vi编辑sysctl.conf文件,增加要修改的内核参数内容,修改的格式为:参数名=值。例如,把net.ipv4.ip_forward的值修改为1,在sysctl.conf中增加下面这行内容:net.ipv4.ip_forward=1 文件修改好后,进行保存。然后使用“sysctl -p 配置文件名”来使配置生效,如果配置文件是默认的,可以不用输配置文件名,即使用“sysctl -p”。 通过文件设置的方式修改的内核参数是在系统重启后将失效(我之前认为修改后的内核参数放在文件中,系统启动的时候会读这个文件,重启后设置应该不会失效。但经过验证,一般会失效,但如果把将默认的boot.sysctl服务打开,所以系统启动时就会执行这个文件的设置)。把我们修改参数的命令写入启动执行脚本文件里/etc/rc.local,这样系统重启后配置就不会失效。 文件方式的好处是内核参数设置的值可以用文件保留下来,调用“sysctl -p”可以使文

摇表、万用表、钳形表的使用方法

摇表(兆欧表),万用表,钳形表的使用方法 电工常用测量仪表有摇表、万用表和钳形电流表,这些仪表在测量时若不注意正确的使用方法或稍有疏忽,不是将表烧坏,就是使被测元件损坏,甚至还危及人身安全,因此,掌握摇表(兆欧表),万用表,钳形表的使用方法。 电工常用测量仪表有摇表、万用表和钳形电流表,这些仪表在测量时若不注意正确的使用方法或稍有疏忽,不是将表烧坏,就是使被测元件损坏,甚至还危及人身安全,因此,掌握常用电工测量仪表的正确使用方法是非常重要的。 1 摇表 摇表又称兆欧表,其用途是测试线路或电气设备的绝缘状况。使用方法及注意事项如下: (1) 首先选用与被测元件电压等级相适应的摇表,对于500V及以下的线路或电气设备,应使用500V或1000V 的摇表。对于500V以上的线路或电气设备,应使用1000V或2500V 的摇表。 (2) 用摇表测试高压设备的绝缘时,应由两人进行。 (3) 测量前必须将被测线路或电气设备的电源全部断开,即不允许带电测绝缘电阻。并且要查明线路或电气设备上无人工作后方可进行。 (4) 摇表使用的表线必须是绝缘线,且不宜采用双股绞合绝缘线,其表线的端部应有绝缘护套;摇表的线路端子“L”应接设备的被测相,接地端子“E”应接设备外壳及设备的非被测相,屏蔽端子“G”应接到保护环或电缆绝缘护层上,以减小绝缘表面泄漏电流对测量造成的误差。 (5) 测量前应对摇表进行开路校检。摇表“L”端与“E”端空载时摇动摇表,其指针应指向“∞”;摇表“L”端与“E”端短接时,摇动摇表其指针应指向“0”。说明摇表功能良好,可以使用。 (6) 测试前必须将被试线路或电气设备接地放电。测试线路时,必须取得对方允许后方可进行。

VMWARE VTSP考题资料(vSphere 存储).

VTSP-CH-Core-Storage VTSP VI 存储 Q: 卷的大小 A: 可小于或大于物理磁盘驱动器的大小 必须与物理磁盘驱动器大小一致 必须比物理磁盘驱动器大255 GB 以上都不是 Q: 以下关于NFS 文件系统的说法哪一种不正确?A: 它要求使用NFS 版本3 或更高版本 它专门为存储ESX 虚拟机而设计 一台ESX 主机最多可连接8 个NFS 数据存储 它可用于为HA 和VCB 配置的虚拟机 Q: 以下哪一选项是HBA 的全球唯一标识符名称? A: 全球通用节点名 全球通用端口名 以上都不是

Q: 以下关于多路径的说法哪些不正确?A: 多路径支持要求用户安装特定的故障切换驱动程序。您必须与阵列供应商联系以获得故障切换驱动程序。 ESX 支持多路径,可维护服务器计算机和存储设备之间的稳定连接 为了支持路径切换,服务器通常使用两个或两个以上可从存储阵列中通过一个或多个交换机访问的HBA 以上都是 Q: ___________为虚拟机内的虚拟磁盘提供存储空间,并存储虚拟机的内容。 A: 群集 数据存储 主机 资源池 Q: 以下哪一文件与虚拟机无关: A: .vmcp .vswp .vmdk .vmss Q:

以下哪一种说法是配置数据存储时建议您使用的最佳做法: A: 每个LUN 对应一个VMFS 卷并仅创建一个VMDK 文件 每个LUN 对应一个VMFS 卷并创建多个VMDK 文件 在每个LUN 中创建多个VMFS 卷,并在每个卷上仅创建一个VMDK 文件 在每个LUN 中创建多个VMFS 卷,并在每个卷上创建多个VMDK 文件 Q: __________是一个群集文件系统,它允许多个物理服务器同时读取或写入至相同的存储设备A: VMFS NFS RDM LUN Q: 您可将光纤通道SAN 和iSCSI SAN 格式化为 A: NFS 卷 RDM VMFS 卷 数据存储 Q:

Linux 内核编译配置选项简介

General setup常规设置 Local versio n - append to kernel release 在内核版本后面加上自定义的版本字符串(小于64字符),可以用"uname -a"命 令看到 Automatically append version information to the versio n string 自动在版本字符串后面添加版本信息,编译时需要有perl以及git仓库支持 Support for paging of anonymous memory (swap) 使用交换分区或者交换文件来做为虚拟内存 System V IPC System V进程间通信(IPC)支持,许多程序需要这个功能.必选,除非你知道自己 在做什么 POSIX Message Queues POSIX消息队列,这是POSIX IPC中的一部分 BSD Process Accounting 将进程的统计信息写入文件的用户级系统调用,主要包括进程的创建时间/创建者/ 内存占用等信息 Export task/process statistics through netlink 通过netlink接口向用户空间导出任务/进程的统计信息,与BSD Process Accounting的不同之处在于这些统计信息在整个任务/进程生存期都是可用的 UTS Namespaces UTS名字空间支持,不确定可以不选 Auditing support 审计支持,某些内核模块(例如SELinux)需要它,只有同时选择其子项才能对系统 调用进行审计 Kernel .config support 把内核的配置信息编译进内核中,以后可以通过scripts/extract-ikconfig脚本来 提取这些信息 Cpuset support 只有含有大量CPU(大于16个)的SMP系统或NUMA(非一致内存访问)系统才需 要它 Kernel->user space relay support (formerly relayfs) 在某些文件系统上(比如debugfs)提供从内核空间向用户空间传递大量数据的接 口

linux内核配置模块编译安装

Linux内核配置编译和加载 Linux内核模块 Linux内核结构非常庞大,包含的组件也非常多,想要把我们需要的部分添加到内核中,有两个方法:直接编译进内核和模块机制 由于直接编译进内核有两个缺点,一是生成的内核过大,二是每次修改内核中功能,就必须重新编译内核,浪费时间。因此我们一般采用模块机制,模块本身不被编译进内核映像,只有在加载之后才会成为内核的一部分,方便了修改调试,节省了编译时间。 配置内核 (1)在drivers目录下创建hello目录存放hello.c源文件 (2)在hello目录下新建Makefile文件和Kconfig文件 Makefile文件内容: obj-y += hello.o //要将hello.c编译得到的hello.o连接进内核 Kconfig文件内容: 允许编译成模块,因此使用了tristate (3)在hello目录的上级目录的Kconfig文件中增加关于新源代码对应项目的编译配置选项 修改即driver目录下的Kconfig文件,添加

source "drivers/hello/Kconfig" //使hello目录下的Kconfig起作用 (4)在hello目录的上级目录的Makefile文件中增加对新源代码的编译条目 修改driver目录下的Makefile文件,添加 obj-$(CONFIG_HELLO_FOR_TEST) += hello/ //使能够被编译命令作用到 (5)命令行输入“make menuconfig”,找到driver device,选择select,发现test menu 已经在配置菜单界面显示出来 (6)选择test menu进入具体的配置,可以选择Y/N/M,这里我选择编译为M,即模块化 (7)保存退出后出现 (8)进入kernels目录中使用“ls -a”查看隐藏文件,发现多出.config隐藏文件,查看.config 文件

史上最全linux内核配置详解

对于每一个配置选项,用户可以回答"y"、"m"或"n"。其中"y"表示将相应特性的支持或设备驱动程序编译进内核;"m"表示将相应特性的支持或设备驱动程序编译成可加载模块,在需要时,可由系统或用户自行加入到内核中去;"n"表示内核不提供相应特性或驱动程序的支持。只有<>才能选择M 1. General setup(通用选项) [*]Prompt for development and/or incomplete code/drivers,设置界面中显示还在开发或者还没有完成的代码与驱动,最好选上,许多设备都需要它才能配置。 [ ]Cross-compiler tool prefix,交叉编译工具前缀,如果你要使用交叉编译工具的话输入相关前缀。默认不使用。嵌入式linux更不需要。 [ ]Local version - append to kernel release,自定义版本,也就是uname -r可以看到的版本,可以自行修改,没多大意义。 [ ]Automatically append version information to the version string,自动生成版本信息。这个选项会自动探测你的内核并且生成相应的版本,使之不会和原先的重复。这需要Perl的支持。由于在编译的命令make-kpkg 中我们会加入- –append-to-version 选项来生成自定义版本,所以这里选N。 Kernel compression mode (LZMA),选择压缩方式。 [ ]Support for paging of anonymous memory (swap),交换分区支持,也就是虚拟内存支持,嵌入式不需要。 [*]System V IPC,为进程提供通信机制,这将使系统中各进程间有交换信息与保持同步的能力。有些程序只有在选Y的情况下才能运行,所以不用考虑,这里一定要选。 [*]POSIX Message Queues,这是POSIX的消息队列,它同样是一种IPC(进程间通讯)。建议你最好将它选上。 [*]BSD Process Accounting,允许进程访问内核,将账户信息写入文件中,主要包括进程的创建时间/创建者/内存占用等信息。可以选上,无所谓。 [*]BSD Process Accounting version 3 file format,选用的话统计信息将会以新的格式(V3)写入,注意这个格式和以前的v0/v1/v2 格式不兼容,选不选无所谓。 [ ]Export task/process statistics through netlink (EXPERIMENTAL),通过通用的网络输出工作/进程的相应数据,和BSD不同的是,这些数据在进程运行的时候就可以通过相关命令访问。和BSD类似,数据将在进程结束时送入用户空间。如果不清楚,选N(实验阶段功能,下同)。 [ ]Auditing support,审计功能,某些内核模块需要它(SELINUX),如果不知道,不用选。 [ ]RCU Subsystem,一个高性能的锁机制RCU 子系统,不懂不了解,按默认就行。 [ ]Kernel .config support,将.config配置信息保存在内核中,选上它及它的子项使得其它用户能从/proc/ config.gz中得到内核的配置,选上,重新配置内核时可以利用已有配置Enable access to .config through /proc/config.gz,上一项的子项,可以通过/proc/ config.gz访问.config配置,上一个选的话,建议选上。 (16)Kernel log buffer size (16 => 64KB, 17 => 128KB) ,内核日志缓存的大小,使用默认值即可。12 => 4 KB,13 => 8 KB,14 => 16 KB单处理器,15 => 32 KB多处理器,16 => 64 KB,17 => 128 KB。 [ ]Control Group support(有子项),使用默认即可,不清楚可以不选。 Example debug cgroup subsystem,cgroup子系统调试例子 Namespace cgroup subsystem,cgroup子系统命名空间 Device controller for cgroups,cgroups设备控制器

相关文档
最新文档