Thierry-2007 Effect of particle size on LiMnPO4 cathodes-jps
雾化吸入文献-1
Deposition and dispersion of aerosols in the airways of the human respiratory tract: the effect ofparticle sizeAuthors: G Scheuch, W StahlhofenSmall volumes of aerosols (boluses) were inspired predominantly into the conducting airways of human lungs with a fast operating valve system, injecting preselected aerosol volumes near the end of a clean air inhalation. Particle recovery and bolus dispersion in the exhaled air after various periods of breathholding were investigated by measuring aerosol number concentration directly in front of the mouth with a laser photometer. Inspired and expired flow rates were measured with a pneumotachograph. The effect of particle size on these measurements has been investigated using aerosol particles with aerodynamic diameters (dae) between 0.9 and 5 microns. For aerosol particles smaller than 2 microns, bolus dispersion increases with increasing periods of breathholding (tb). After reaching a maximum, dispersion decreases with even longer tb. An increase in particle size yields a smaller increase in dispersion during the first seconds of breathholding while it is not changed significantly without breathhold. Particle losses during inhalation and exhalation increases with particle size. However, with increasing periods of breathholding, the losses of the smaller particles (less than 1.5 microns) were found to be much higher than expected theoretically, implying particle losses by sedimentation in the same airway structures. The small aerosol particles are deposited in smaller airways than bigger particles. These observations can be explained by cardiogenic mixing during periods of breathholding by pulsatile flow oscillations and confirm measurements with enhanced heart rate as described in an earlier paper. Small particles with restricted settling velocities remained longer in an airborne state in the airways and this leads to a more efficient cardiogenic mixing. Experimental lung research. 18(3):343-58.ISSN: 0190-2148Comparison of three commercial ultrasonic nebulizersAuthors: R K Katial, C Reisner, A Buchmeier, B B Bartelson, H S NelsonBACKGROUND: The clinical acceptance of the initial ultrasonic nebulizers was impeded by their production of significant quantities of droplets larger than the respirable range that could have resulted in poor pulmonary deposition of nebulized medications. Subsequent modifications in the design of ultrasonic nebulizers have occurred. Overall nebulizer performance characteristics of the newer ultrasonic devices have not been evaluated. OBJECTIVE: Three commercially available ultrasonic nebulizers (DeVilbiss-Pulmosonic, Omron-Microair, Rh.nePoulenc-Rorer-Fisoneb) were studied to compare the aerosol output characteristics. METHODS: The parameters studied were total volume output (TVO), time to nebulize total output (TTO), percent of droplets with volume diameters in the respirable range (PDVRR, 1 to 5 microm), albuterol concentration during nebulization, and the total drug delivered. All nebulizers were filled with 2.5 mL of saline and 0.5 mL of albuterol nebulizer solution. Three units from each manufacturer, each from a different lot, were evaluated in duplicate. RESULTS: The nebulizer with the largest volume output was the Omron (mean 2.94 mL), which also demonstrated the longest nebulization time (mean 10.3 min). The DeVilbiss and Rh.ne Poulenc-Rorer units delivered smaller volumes (mean 2.5 mL, 2.4 mL, respectively) but nebulized more rapidly (mean 2.21 min, 3.54 min, respectively). The Omron nebulizer generated the highest PDVRR with a mean of 38%. The DeVilbiss had a mean PDVRR of 16% and the Rh.ne Poulenc-Rorer a mean PDVRR of 21%. The majority of droplets from all three machines had a volume diameter smaller than the respirable range, ie, in the 0.5 to 1.0 microm range (Omron-60%, DeVilbiss-83%, Rh.ne Poulenc-Rorer-79%). For all three nebulizers there appeared to be no concentrating or diluting effect during nebulization implying that equal quantities of albuterol and diluent were delivered. The Rh.ne Poulenc-Rorer units demonstrated the greatest unit-to-unit variability with respect to TVO while the Omron units demonstrated the greatest unit to unit variability with respect to TTO. CONCLUSION: We conclude that several improvements in the design of ultrasonic nebulizers have resulted in the reduction of the size of the droplets generated. Our evaluation of the three commercially available ultrasonic nebulizers revealed that the majority of droplets generated were within or below the respirable range. There was no concentrating or diluting effect during nebulization for all three nebulizers. The output characteristics of the three devices differ and this will effect the delivery time as well as amount of drug delivered to the lungs.Annals of allergy, asthma & immunology : official publication of the American College of Allergy, Asthma, &Immunology. 01/03/2000; 84(2):255-61.ISSN: 1081-1206The function profile of compressed-air and ultrasonic nebulizers Authors: Hsin-Lin Wu, Yung-Zen Lin, W ei-Fong Wu, Fu-Yuan HuangIn order to study the detailed function of two kinds of nebulizers commonly used in clinical asthma treatment, compressed-air and ultrasonic, this study was conducted. At the beginning, various flow rates were adjusted, paired with different volumes of solutions in the container. The changes of temperature, pH, and osmolality during the course of nebulization were examined. Normal saline, terbutaline, and fenoterol solutions were used as the nebulized solutions. The study was performed in an environment in ambient temperature around 20 degrees C and relative humidity around 70%. The results showed a minimal 6 L/min flow rate was required to nebulize the solution when using the compressed-air nebulizer. The dead volume was about 0.8 ml for compressed-air and 8.5 ml for the ultrasonic nebulizer. When using the compressed-air nebulizer, the temperature, both in the solution and at the mouthpiece site, dropped gradually. On the contrary, the temperatures at both sites increased a little bit when using the ultrasonic nebulizer. The pH values of pure terbutaline and fenoterol nebulized solutions were acidic (3.58 and 3.00 respectively). The osmolality of terbutaline and fenoterol nebulized solutions were isotonic. The osmolality increased gradually during the course of nebulization, to a greater extent in the compressed-air nebulizer. In conclusion, both types of nebulizers have their special features. The ultrasonic nebulizer displays less extent in change of temperature and osmolality during nebulization and is expected to be a better device in treating asthmatic patients in terms of lesser effect on cooling and changing the osmolality of airway mucosa. Acta paediatrica Taiwanica = Taiwan er ke yi xue hui za zhi.44(5):264-8.ISSN: 1608-8115。
阿贡实验室RietveldRefinementwithGSAS
5
Errors & parameters?
position – lattice parameters, zero point (not common)
- other systematic effects – sample shift/offset shape – profile coefficients (GU, GV, GW, LX, LY, etc. in GSAS) intensity – crystal structure (atom positions & thermal parameters)
9
So how does Rietveld refinement work?
Rietveld Minimize
MR w(Io Ic )2
Io
Exact overlaps - symmetry
Incomplete overlaps
SIc
Residuals:
Rwp
w(I o Ic )2
wI
c2
Least-squares cycles
False minimum
True minimum – “global” minimum
parameter c2 surface shape depends on parameter suite
3
Fluoroapatite start – add model (1st choose lattice/sp. grp.)
c
5i
ig
i
i1
18
Detection-of-QTLs-with-Additive-Effects
Agricultural Sciences in China 2009, 8(9): 1039-1045September 2009© 2009, CAAS. All rights reserved. Published by Elsevier Ltd.Detection of QTLs with Additive Effects, Epistatic Effects, and QTL ×Environment Interactions for Zeleny Sedimentation Value Using a Doubled Haploid Population in Cultivated WheatZHAO Liang, LIU Bin, ZHANG Kun-pu, TIAN Ji-chun and DENG Zhi-yingState Key Laboratory of Crop Biology, Group of Quality Wheat Breeding, Shandong Agricultural University, Tai’an 271018, P.R.ChinaAbstractIn order to understand the genetic basis for Zeleny sedimentation value (ZSV) of wheat, a doubled haploid (DH) population Huapei 3×Yumai 57 (Yumai 57 is superior to Huapei 3 for ZSV), and a linkage map consisting of 323 marker loci were used to search QTLs for ZSV. This program was based on mixed linear models and allowed simultaneous mapping of additive effect QTLs, epistatic QTLs, and QTL×environment interactions (QEs). The DH population and the parents were evaluated for ZSV in three field trials. Mapping analysis produced a total of 8 QTLs and 2 QEs for ZSV with a single QTL explaining 0.64-14.39% of phenotypic variations. Four additive QTLs, 4 pairs of epistatic QTLs, and two QEs collectively explained 46.11% of the phenotypic variation (PVE). This study provided a precise location of ZSV gene within the Xwmc 93 and GluD1 interval, which was designated as Qzsv-1D. The information obtained in this study should be useful for manipulating the QTLs for ZSV by marker assisted selection (MAS) in wheat breeding programs.Key words: doubled haploid population, Zeleny sedimentation value, quantitative trait loci (QTLs), wheat (Triticum aestivum L.)INTRODUCTIONThe Zeleny sedimentation value (ZSV) has been provento be useful in wheat breeding programs for the esti-mation of wheat eating and cooking quality (Mesdag1964; Kne et al. 1993; Liu et al. 2003; He et al.2004; Zhang et al. 2005; Özberk et al. 2006; Ozturket al. 2008). There is a positive correlation betweensedimentation volume and gluten strength or loaf volume.The ZSV method is often used as a screening test inwheat breeding. Mesdag (1964) showed that the valueof ZSV is a measure for the quantity and quality of thegluten. Because the baking value of wheat flour is largelydetermined by these components, the ZSV is also con-sidered as a useful predictor for the baking value. LiuReceived 3 December, 2008 Accepted 9 April, 2009Correspondence TIAN Ji-chun, Professor, Tel/Fax: +86-538-8242040, E-mail: jctian9666@et al. (2003) detected that the associations betweenZSV and DWCN’s (dry white Chinese noodle) appear-ance and taste also fit quadratic regression modelsignificantly. The gluten quality-related parameter ofsedimentation value was significantly associated withpan bread quality score (He et al. 2004). Özberk et al.(2006) found that the only quality analyses showingsignificant correlations with market price were Zelenysedimentation value and hectolitre weights (kg hL-1).Ozturk et al. (2008) reported that the cookie diametergave highly significant correlations with ZSV.The advent and utilization of molecular markers hasprovided powerful tools for elucidating the genetic ba-sis of quantitatively inherited traits. However, only afew studies have reported genetic loci that influenceZSV in wheat (Rousset et al. 2001; Kunert et al. 2007;1040ZHAO Liang et al.Sun et al. 2008). Rousset et al. (2001) reported that one strong QTL for ZSV was mapped on the long arm of chromosome 1A around Glu-A1. A distally located QTL for ZSV was mapped on chromosome arm 1BS, centered on the Gli-B1/Glu-B3 region. And a major QTL for ZSV, clearly corresponding to the Glu-D1 locus, was detected on chromosome arm 1DL. Kunert et al. (2007) found four putative QTLs for ZSV. Sun et al. (2008) identified three QTLs for ZSV in a F14 RIL derived from the cross between Chuan 35050 and Shannong 483.Additive effect QTLs were first identified and epi-static interactions among these additive effect QTLs were then estimated (Zanetti et al. 2001). However, this approach usually leaves out many QTLs that may have no additive effects but influence the trait only through epistatic interactions or QTL×environment in-teractions (QEs) (Ma et al. 2005, 2007; Rebetzke et al. 2007). Additive effect QTLs, epistatic QTLs, and QEs were detected using two-locus analyses in both the populations (Kulwal et al. 2005). Sometimes QTLs involved in such interactions contribute substantially to the total variation of a quantitative trait, and therefore should not be ignored. Further experimentation is needed to clarify whether the traits are also affected by epistatic and environment, and to dissect the genotype ×environment interaction effects at the molecular level. In this study, QTLs for ZSV were investigated based on the mixed linear model in a DH population across environments. The objective of this study was to com-prehensively characterize the genetic basis for ZSV of wheat in order to facilitate the future breeding of high-quality wheat varieties.MATERIALS AND METHODSMaterialsA population of 168 DH lines was produced from the cross between two Chinese wheat cultivars Huapei 3 (Hp3)/Yumai 57 (Ym57) and was used for the con-struction of a linkage map. The DH population and parents were kindly provided by Professor Yanhai, Henan Academy of Agricultural Sciences, Zhengzhou, China. Hp3 and Ym57 were registered by Henan Prov-ince of China in 2006 (Hai and Kang 2007) and by the state (China) in 2003 (Guo et al. 2004), respectively. The parents, planted over a large area in the Huang-Huai wheat region in China, differ in several agronomi-cally important traits as well as baking quality traits (Guo et al. 2004; Hai and Kang 2007).The field trials were conducted in three environments, at Tai’an (36.18°N, 117.13°E), Shandong Province, China, in 2005 and 2006, and at Suzhou (31.32°N, 120.62°E), Anhui Province, China, in 2006. The ex-perimental design followed a completely randomized block design with two replications at each location. In autumn 2005, all lines and parental lines were grown in 2 m long by three-row plots (25 cm apart); in autumn 2006, the lines were grown in 2 m long by four-row plots (25 cm apart). Suzhou and Tai’an differ in cli-mate and soil conditions. In Tai’an, there were differ-ences in temperature and soil conditions between the years 2005 and 2006. During the growing season, man-agement was in accordance with the local practice. The lines were harvested individually at maturity to prevent yield loss from over-ripening. Harvested grain samples were cleaned prior to conditioning and flour milling was performed in a mill (Quadrumat Senior, Brabender, Germany) to flour extraction rates of around 70%. Prior to milling, the hard, medium hard (mixtures of hard and soft wheat) and soft wheats were tempered to around 14, 15, and 16% moisture contents, respectively.Measurements of ZSVZeleny sedimentation volume was determined using AACC method 56-61A.Construction of the genetic linkage mapA genetic linkage map of DH population with 323 markers, including 284 SSR, 37 ESTs loci, 1 ISSR loci and 1 HMW-GS loci, was constructed. This linkage map covered a total length of 2485.7 cM with an aver-age distance of 7.67 cM between adjacent markers. Thirteen markers remained unlinked. These markers formed 24 linkage groups at LOD 4.0. The chromo-somal locations and the orders of the markers in the map were in accordance with the one reported for Triti-cum aestivum L. (Somers et al. 2004). The recom-mended map distance for genome wide QTL scanningDetection of QTLs with Additive Effects, Epistatic Effects, and QTL×Environment Interactions for Zeleny Sedimentation1041 was an interval length less than 10 cM (Doerge 2002).Thus the map was suitable for genome-wide QTL scan-ning in this study.Statistical analysisAnalysis of variance (ANOVA) was carried out usingSPSS ver. 13.0 (SPSS, Chicago, USA). QTLs withadditive effects and epistatic effects as well as QEs inthe DH population were mapped by the softwareQTLNetwork ver. 2.0 (Yang and Zhu 2005) based on amixed linear model (Wang et al. 1999). Composite in-terval analysis was undertaken using forward-backwardstepwise multiple linear regression with a probabilityinto and out of the model of 0.05 and window size setat 10 cM. Significant thresholds for QTL detectionwere calculated for each data set using 1000 permuta-tions and a genome-wide error rate of 0.10 (suggestive)and 0.05 (significant). The final genetic model incor-porated significant additive effects and epistatic effectsas well as their environmental interactions.RESULTSPhenotypic variation for DH lines and parentsAs is shown in Fig.1, ZSV of Ym57 showed highervalues than ZSV of Hp3; the means of the ZSV fellbetween the two parent’s values. It expressed the ex-istence of the large transgressive segregation. ZSV seg-regated continuously and approximately fit normal dis-tributions with absolute values of both skewness andkurtosis less than 1.0, indicating that this trait was suit-able for QTL mapping.QTLs with additive effects and additive×environment (AE) interactionsFour QTLs with significant additive effects were iden-tified on chromosomes 1B, 1D, 5A, and 5D (Table 1and Fig.2). These QTLs explained from 2.66 to14.39% of the phenotypic variance. The Qzsv-1B had the most significant effect, accounting for 14.39% of the phenotypic variance. The Ym57 alleles at three loci, Qzsv-1B,Qzsv-1D, and Qzsv-5D, increased Fig. 1 Frequency distributions of ZSV in 168 DH lines derived from a cross of Hp3×Ym57 evaluated at three environments in the 2005 and 2006 cropping seasons. The means of trait values for the DH lines and both parents are indicated by arrows. Several statistics for the traits in the DH lines are shown on the right of each plot.Zeleny sedimentation volume (mL)2006 in SuzhouZeleny sedimentation volume (mL)2006 in Tai’anZeleny sedimentation volume (mL)2005 in Tai’anMean: 24.39SD: 5.45Range: 12.00-39.00Skewness: 0.171Kurtosis: -0.153 252015105No.ofDHlinesDH linesYm57Hp315.0020.0025.0030.0035.0040.00DH linesYm57Hp320.0030.0040.0050.0060.00252015105No.ofDHlines30DH linesYm57Hp320.0030.0040.002015105No.ofDHlinesMean: 24.39SD: 5.45Range: 12.00-39.00Skewness: 0.171Kurtosis: -0.153Mean: 24.39SD: 5.45Range: 12.00-39.00Skewness: 0.171Kurtosis: -0.1531042ZHAO Liang et al.Table 1 Estimated additive effects and additive ×environment (AE) interactions of QTLs for ZSV at three environments in the 2005 and 2006 cropping seasonsQTL Flanking-marker 1)Position (cM)2)F -value P A 3)H 2 (A, %)4)AE 1H 2 (AE 1, %)5)AE 2H 2 (AE 2, %)AE 3H 2 (AE 3, %)Qzsv -1B Xwmc412.2-Xcfe023.236.425.220.000-2.5214.39------Qzsv -1D Xwmc93-GluD161.915.910.000-1.988.93------Qzsv -5A Xbarc358.2-Xgwm18638.18.100.000 1.08 2.66------Qzsv -5DXcfd101-Xbarc32060.612.690.000-1.203.25---1.042.44--1)Flanking marker, the interval of F peak value for QTL. The same as below.2)Position, the location of F peak value for QTL in “Flanking marker”. The same as below.3)Additive effects, a positive value indicates that the allele from Hp3 increased ZSV, a negative value indicates that the allele from Ym57 increased ZSV.4)H 2(A, %) indicates the contribution explained by putative additive QTL.5)H 2(AE 1, %) indicates the contribution explained by additive QTL ×environment 1 interaction. E 1, Tai’an 2005; E 2, Tai’an 2006; E 3, Suzhou 2006.Fig. 2 A genetic linkage map of wheat showing mapping QTLs with additive effects, epistatic effects, AE, and AAE for ZSV.1A 1B 1D 2A 3A5A 5D 7A 7DLocus involved in AELocus involved in additive effects Locus involved in epistasisLocus involved in AAEDetection of QTLs with Additive Effects, Epistatic Effects, and QTL ×Environment Interactions for Zeleny Sedimentation 1043ZSV by 2.52, 1.98, and 1.20 mL, respectively, owing to additive effects. The Hp3 allele increased ZSV at the Qzsv -5A by 1.08 mL, accounting for 2.66% of the phe-notypic variance. This suggested that alleles, which increased ZSV, were dispersed within the two parents,resulting in small differences of phenotypic values be-tween the parents and transgressive segregants among the DH population. The total additive QTLs detected for ZSV accounted for 29.23% of the phenotypic variance.One additive effect was involved in AE interactions (Table 1 and Fig.2). The Ym57 alleles at one locus,Qzsv -5D , increased the ZSV by 1.04 mL with corre-spondingly contributing 2.44% of the phenotypic variance.QTLs with epistasis effects and epistasis ×environment (AAE) interactionsFour pairs of epistatic QTLs were identified for ZSV,and were located on chromosomes 1A, 2A, 3A, 7A and 7D (Table 2 and Fig.2). These QTLs had correspond-ing contributions ranging from 0.64 to 6.79%. One pair of epistasis, occurring between the loci Qzsv -2A /Qzsv -7A , had the largest effect, which contributed ZSV of 1.73 mL and accounted for 6.79% of the phenotypic variance. The four pairs of epistatic QTLs explained 12.11% of the phenotypic variance. All the epistatic effects were non-main-effect QTLs.One pair of epistatic QTL was detected in AAE in-teractions for ZSV (Table 2 and Fig.2). The AAE ef-fects explained 2.33% of the phenotypic variance and this QTL, Qzsv3A.2/Qzsv7D.1, increased ZSV by 1.01mL owing to AAE effects, simultaneously the positive value means that the parent-type effect is greater than the recombinant-type effect.DISCUSSIONEpistatic effects and QTL ×environment interactions were important genetic basis for ZSV in wheatEpistasis, as an important genetic basis for complex traits, has been well demonstrated in recent QTL map-ping studies (Cao et al . 2001; Fan et al . 2005; Ma et al .2005, 2007). Ma et al . (2005) provided a strong evi-dence for the presence of epistatic effects on dough rheological properties in a wheat DH population. In the present study, four pairs of QTLs with epistatic ef-fects were detected for ZSV in three environments (Table 2 and Fig.2). The four pairs of epistatic QTLs explained 12.11% of the phenotypic variance.ZSV was predominantly influenced by the effects of genotype (Zhang et al . 2004, 2005), and in the present study, only one AE interaction and one AAE interaction were found. It is suggested that QTL ×environment interactions just play a minor role, but QTL ×environment interactions should not be ignored.ZSV and subunits of high molecular weight gluteninsSubunits of high molecular weight glutenins strongly influence wheat bread making quality. This study pro-vided a precise location of ZSV gene within the Xwmc 93 and GluD1 interval, which was designated Qzsv -1D and was located in the central region of a 2 cM interval.Also Rousset et al . (2001) detected a major QTL for sedimentation volume on 1DL, clearly corresponding to the Glu -D1 locus. Kunert et al . (2007) found that the SSR marker Xgwm642 on 1DL identified a QTLTable 2 Estimated epistatic effects and epistasis ×environment (AAE) interactions of QTLs for ZSV at three environments in the 2005 and 2006 cropping seasonsPosition Position H 2H 2H 2H 2(cM)(cM)(AA, %)2)(AAE 1, %)3)(AAE 2, %)(AAE 3, %)Qzsv -1A Xwmc278-Xbarc120.156.3Qzsv -3A.1Xbarc1177-Xbarc276.2196.3-0.94 1.99------Qzsv -2A Xgwm636-Xcfe6729.1Qzsv -7A Xbarc259-Xwmc59653.7-1.73 6.79------Qzsv -3A.2Xcfa2193-Xgwm155152.7Qzsv -7D.1Xcfd175-Xwmc14181.5-1.09 2.69 1.01 2.33----Qzsv -3A.2Xcfa2193-Xgwm155152.7Qzsv -7D.2Xgdm67-Xwmc634161.5-0.530.64------1)The epistatic effect. A positive value means that the parent-type effect is greater than the recombinant-type effect, and the negative value means that the parent-type effect is less than the recombinant-type effect.2)H 2 (AA, %) indicates the contribution explained by putative epistatic QTL.3)H 2 (AAE 1, %) indicates the contribution explained by epistatic QTL ×environment 1 interaction. E 1, Tai’an 2005; E 2, Tai’an 2006; E 3, Suzhou 2006.QTL Flanking-marker QTL Flanking-markerAA 1)AAE 1AAE 2AAE 31044ZHAO Liang et al. for ZSV. The position indicates an influence of theGlu-D1 locus. And a major QTL, clearly correspond-ing to the Glu-D1 locus, was detected on chromosomearm 1DL. Correlation coefficient between Glu-1 scoreand sedimentation values was significant (r=0.553).There were significant correlations between sedimen-tation values and Glu-lAa,Glu-1Ac,Glu-Ba, and Glu-1Bcalleles, respectively (Kne et al. 1993). Thesedimentation values showed statistically significantassociations with the status of the Glu-A1 locus(Witkowski et al. 2008).In this study, the Qzsv-1D increased ZSV by 1.98mL, correspondingly contributing 8.93% of the pheno-typic variance. Barro et al. (2003) found that HMW-GS 1Ax1 increased the sedimentation value. In contrast,HMW-GS 1Dx5 drastically decreased in sedimentationvalue.In summary, four additive QTLs, four pairs of epi-static QTLs, and two QEs were detected for ZSV in168 DH lines derived from a cross Hp3×Ym57. Onemajor QTL,Qzsv-1B, was closely linked to Xwmc412.20.2cM and could account for 14.39% of the phenotypicvariation without any influence from the environment.Therefore, the Qzsv-1B could be used in MAS in wheatbreeding programs. The results showed that both ad-ditive and epistatic effects were important as a geneticbasis for ZSV, and were also sometimes subject to en-vironmental modifications.AcknowledgementsThis work was supported by the National Basic Re-search Program of China (2009CB118301), the NationalHigh-Tech Research and Development (863) Programof China (2006AA100101 and 2006AA10Z1E9), andthe Doctor Foundation of Shandong AgriculturalUniversity, China (23023). Thanks Prof. Chuck Walker,University of Kansas State University, USA, for hiskindly constructive advice on the language editing ofthe manuscript.ReferencesBarro F, Barceló P, Lazzeri P A, Shewry P R, Ballesteros J,Martín A. 2003. Functional properties of flours from fieldgrown transgenic wheat lines expressing the HMW gluteninsubunit 1Ax1 and 1Dx5 genes. Molecular Breeding,12,223-229.Cao G, Zhu J, He C, Gao Y, Yan J, Wu P. 2001. Impact ofepistasis and QTL×environment interaction on thedevelopmental behavior of plant height in rice (Oryza sativaL.). Theoretical and Applied Genetics,103, 153-160.Doerge R W. 2002. Multifactorial genetics: Mapping and analysisof quantitative trait loci in experimental populations. NatureReviews,3, 43-52.Fan C C, Yu X Q, Xing Y Z, Xu C G, Luo L J, Zhang Q F. 2005.The main effects, epistatic effects and environmentalinteractions of QTLs on the cooking and eating quality ofrice in a doubled-haploid line population. Theoretical andApplied Genetics,110, 1445-1452.Guo C Q, Bai Z A, Liao P A, Jin W K. 2004. New high qualityand yield wheat variety Yumai 57. China Seed Industry,4, 54(in Chinese)Hai Y, Kang M H. 2007. Breeding of a new wheat vatiety Huapei 3with high yield and early maturing. Henan AgriculturalSciences, 5, 36-37. (in Chinese)He Z H, Yang J, Zhang Y, Quail K J, Peña R J. 2004. Pan breadand dry white Chinese noodle quality in Chinese winterwheats.Euphytica,139, 257-267.,G, D. 1993. Allelic variationat Glu-1 loci in some Yugoslav wheat cultivars. Euphytica,69,89-94.Kulwal P, Kumar N, Kumar A, Balyan H S, Gupta P K. 2005.Gene networks in hexaploid wheat: interacting quantitativetrait loci for grain protein content. Functional & IntegrativeGenomics,5, 254-259.Kunert A, Naz A A, Oliver D, Pillen K, Léon J. 2007. AB-QTLanalysis in winter wheat: I. Synthetic hexaploid wheat(T.turgidum ssp. dicoccoides × T. tauschii) as a source offavourable alleles for milling and baking quality traits.Theoretical and Applied Genetics,115, 683-695.Liu J J, He Z H, Zhao Z D, Peña R J, Rajaram S. 2003. Wheatquality traits and quality parameters of cooked dry whiteChinese noodles. Euphytica,131, 147-154.Ma W, Appels R, Bekes F, Larroque O, Morell M K, Gale K R.2005. Genetic characterisation of dough rheological propertiesin a wheat doubled haploid population: additive genetic effectsand epistatic interactions. Theoretical and Applied Genetics,111, 410-422.Ma X Q, Tang J H, Teng W T, Yan J B, Meng Y J, Li J S. 2007.Epistatic interaction is an important genetic basis of grainyield and its components in maize. Molecular Breeding,20,41-51.Mesdag J. 1964. in the protein content of wheat and its influenceon the sedimentation value and the baking quality. Euphytica,13, 250-261.Özberk I, Kýlýç H, Atlý A, Özberk F, Karlý B. 2006. Selectionof wheat based on economic returns per unit area. Euphytica,Detection of QTLs with Additive Effects, Epistatic Effects, and QTL×Environment Interactions for Zeleny Sedimentation1045152, 235-245.Ozturk S, Kahraman K, Tiftik B, Koksel H. 2008. Predicting the cookie quality of flours by using Mixolab. European Food Research and Technology,227, 1549-1554.Rebetzke G J, Ellis M H, Bonnett D G, Richards R A. 2007.Molecular mapping of genes for Coleoptile growth in bread wheat (Triticum aestivum L.). Theoretical and Applied Genetics,114, 1173-1183.Rousset M, Brabant P, Kota R S, Dubcovsky J, Dvorak J. 2001.Use of recombinant substitution lines for gene mapping and QTL analysis of bread making quality in wheat. Euphytica, 119,81-87.Somers D J, Isaac P, Edwards K. 2004. A high-density microsatellite consensus map for bread wheat (Triticum aestivum L.). Theoretical and Applied Genetics,109, 1105-1114.Sun H Y, Lu J H, Fan Y D, Zhao Y, Kong F, Li R J, Wang H G, Li S S. 2008. Quantitative trait loci (QTLs) for quality traits related to protein and starch in wheat. Progress in Natural Science,18, 825-831.Wang D L, Zhu J, Li Z K, Paterson A H. 1999. Mapping QTLswith epistatic effects and QTL × environment interactions by mixed linear model approaches. Theoretical and Applied Genetics,99, 1255-1264.Witkowski E, Waga J, Witkowska K, Rapacz M, Gut M, Bielawska A, Luber H, Lukaszewski A J. 2008. Association between frost tolerance and the alleles of high molecular weight glutenin subunits present in Polish winter wheats. Euphytica, 159,377-384.Yang J, Zhu J. 2005. Methods for predicting superior genotypes in multiple environments based on QTL effects. Theoretical and Applied Genetics,110, 1268-1274.Zanetti S, Winzeler M, Feuillet C, Keller B, Messmer M. 2001.Genetic analysis of bread-making quality in wheat and spelt.Plant Breeding,120, 13-19.Zhang Y, He Z H, Guo Y Y, Zhang A M, Maarten V G.2004.Effect of environment and genotype on bread-making quality of spring-sown spring wheat cultivars in China. Euphytica, 139, 75-83.Zhang Y, Zhang Y, He Z H, Ye G Y. 2005. Milling quality and protein properties of autumn-sown Chinese wheats evaluated through multi-location trials. Euphytica,143,209-222.(Edited by ZHANG Yi-min)。
基于粒子群算法的多次反射飞行时间质量分析器电压优化
!第!!卷第"期质谱学报#$%&!!!'$&" !!()(*年+月,$-./0%$1234/5657066895:;.$<5;.=8$:45;=859&()(*基于粒子群算法的多次反射飞行时间质量分析器电压优化黄!奇> 任!熠> ( 陈政阁> 陈剑松* 洪!义(梁!欣* 李!梅> ! 黄正旭> ! 周!振> !!>&暨南大学质谱仪器与大气环境研究所"广东广州!">)M*(#(&广州禾信仪器股份有限公司"广东广州!">)"*)#*&广东省麦思科学仪器创新研究院"广东广州!">)?))#!&广东省大气污染在线源解析系统工程技术研究中心"广东广州!">)M*($摘要 多次反射飞行时间!7B D C I^$是一种新型的质量分析器"常用于分析短寿命离子%分离同重元素和存储离子&随着使用需求的增加"提高7B D C I^质量分析器的分辨能力越来越重要&然而"7B D C I^质量分析器电压参数的优化涉及高维度%高精细度和非线性的问题"很难用解析方法得到最优参数&本研究提出了一种基于粒子群!e8I$算法的7B D C I^质量分析器电压参数优化方法"并对粒子群算法进行改进&在8S7S I'离子光学仿真平台上对优化方法进行测试"比较了标准粒子群算法和改进粒子群算法的优化结果&结果表明"改进粒子群算法能够获得超过N>))))的极限质量分辨率"相比标准粒子群算法有更好的性能&该方法具有操作简单%优化速度快%求解效果好等优点"可为7B D C I^质量分析器的电压优化提供方法参考"从而提高7B D C I^质量分析器的开发效率&关键词 多次反射飞行时间质量分析器#粒子群算法!e8I$#电压优化#离子光学模拟中图分类号 I M"?&M*文献标志码 A文章编号 >))!D(++?!()(*$)"D)M M?D)+!"# >)&?"*N'P9Q R&()(*&))(!W"'/&90P7/#4#N&/#"%"*3K'/#7'020*'0./#"%:#40"*I'#9,/ 3&))$%&'(N0-M&)0!"%T&-/#.'0>@&-4P7/#4#N&/#"%$'9"-#/,4GT A'W F4>"B`'V4>"("2G`'Y35/X D X5>"2G`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基于电喷雾解吸电离技术的在线大气有机气溶胶超高质量分辨质谱仪项目!(>*>+))($本文通信作者任熠Copyright©博看网. All Rights Reserved.-%;.0D34X3<066.56$%-;4$/0/Z306R55/-65Z0;4/6;4;-;4$/6"6-:306G5%<3$%;P25/;.5 1$.G50\=S$/B5650.:3!W8S$"`-.$950/I.X0/4P0;4$/1$.'-:%50.B5650.:3!2`B'$ 0/ZG4X3`/5.X=A::5%5.0;$.B5650.:3I.X0/4P0;4$/!O`O$;$<506-.563$.;D%4\5Z4$/ <06656"6590.0;546$R0.0/Z6;$.54$/6&A6;35Z5<0/Z1$.-654/:.5064/X"4;46R5:$<4/X 4<9$.;0/;;$4<9.$\5;35.56$%\4/X9$[5.$17B D C I^<0660/0%=P5.6&G$[5\5.";35 $9;4<4P0;4$/$1;35\$%;0X590.0<5;5.6$17B D C I^<0660/0%=P5.646034X3D Z4<5/64$/D 0%"34X3%=.514/5Z0/Z/$/D%4/50.9.$R%5<"[34:346Z4114:-%;;$6$%\5$9;4<0%%=R=0/0%=;D 4:0%<5;3$Z&S/;3466;-Z="090.;4:%56[0.<$9;4<4P0;4$/!e8I$0%X$.4;3<D R065Z<5;3D $Z1$.$9;4<4P4/X;35\$%;0X590.0<5;5.6$17B D C I^<0660/0%=P5.6[069.$9$65Z&C35 <5;3$Z-65Z0/4<9.$\5Z90.;4:%56[0.<$9;4<4P0;4$/!S e8I$099.$0:3[4;304/5.;40 [54X3;Z5:0=6;.0;5X=&C35$9;4<4P0;4$/<5;3$Z[06;56;5Z$/;358S7S I'4$/$9;4:6 64<-%0;4$/9%0;1$.<&2$/64Z5.4/X>**26H4$/[4;3O K>c"]5#" `K N c"5#" J K 0K ><<" KK>c"<.0Z"0<066.56$%\4/X9$[5.$\5.N c>i>)"[060:345\5Z[35/ &/K)/60/Z0<066.56$%\4/X9$[5.$\5."c)i>)"[060:345\5Z[35/&/K()/6&S e8I $9;4<4P5Z;35R56;.56-%;6;$0:345\5;35(/Z$.Z5.1$:-6$1;4<5[4;3.5695:;;$5/5.X= 1$.;357B D C I^<0660/0%=P5."0/Z;35Z5\40;4$/$1;354$/f630%1D;-./;4<5$11%4X3; [06[4;34/>c*i>)J M&S/();4<565Q95.4<5/;6"S e8I4<9.$\5Z;35<0Q4<-<.56-%;6 R=**b";350\5.0X5.56-%;6R=*"b0/Z;356;0/Z0.ZZ5\40;4$/R=(+b:$<90.5Z[4;3 e8I$9;4<4P0;4$/"9.$\4Z4/X R5;;5.6$%-;4$/_-0%4;=0/Z6;0R4%4;=&S e8I f6%4/50.Z5:0= 6;.0;5X=5115:;4\5%=:$/;.$%%5Z;35.5Z-:;4$/$1;35\$%;0X5-9Z0;56;5964P50/Z[060R%5 ;$<55;;35X%$R0%650.:30/Z.514/5<5/;$1;357B D C I^<0660/0%=P5.\$%;0X590.0<5D ;5.$9;4<4P0;4$/9.$R%5<&S;30Z X$$Z:$/\5.X5/:50/Z:$/\5.X5/:56955Z&C346[$.] 9.$\4Z5Z0106;0/Z5115:;4\5<5;3$Z1$.$9;4<4P4/X;35\$%;0X590.0<5;5.6$1;35 7B D C I^<0660/0%=P5.0/Z35%95Z;$4<9.$\5;3595.1$.<0/:5$1;3460/0%=P5.&C35 .56-%;663$[5Z;30;;35S e8I460R%5;$$R;04/0%4<4;4/X<066.56$%-;4$/$1<$.5;30/ N>))))"[34:3306R5;;5.95.1$.<0/:5:$<90.5Z[4;3;35e8I&C35<5;3$Z306;35 0Z\0/;0X56$164<9%5$95.0;4$/"106;$9;4<4P0;4$/0/ZR5;;5.6$%-;4$/"[34:3:0/9.$\4Z5 0<5;3$Z.515.5/:51$.\$%;0X5$9;4<4P0;4$/$17B D C I^<0660/0%=P5.0/Z;3-64<9.$\5 ;35Z5\5%$9<5/;5114:45/:=$17B D C I^<0660/0%=P5.&0(@"-!)*<-%;49%5.51%5:;4$/;4<5D$1D1%4X3;<0660/0%=P5.#90.;4:%56[0.<$9;4<4P0;4$/ !e8I$#$9;4<4P0;4$/$1\$%;0X5#4$/$9;4:664<-%0;4$/!!飞行时间质谱!C I^D78$是一种常用的质谱分析技术(>D()"具有灵敏度高%操作简单等优点"被广泛用于生物医药(*)%环境科学(!)%食品科学(")等领域&多次反射飞行时间!7B D C I^$作为一种新型的质量分析器"通过在(组无网反射镜之间形成的非线性静电场中约束离子"延长离子的飞行距离和飞行时间(M)"以获得超高的质量分辨率!%>i>)"$"现已被亥姆霍兹重离子研究中心!W8S$(D N)%欧洲核子研究中心!2`B'$(+D>))和高能加速器研究机构!O`O$(>>D>()等机构用于测量短寿命离子质量(>*)%分离同重核素(+">!D>")以及存储离子(>M D>?)等&质量分辨率受电压参数的影响"7B D C I^质量分析器具有以下特点*>$>)多个电极#($电压敏感!变化>)J""质量分辨率下降")b$#*$电压组合规律不明显!非线性静电场$&因此"电压优化工作非常繁琐"使用人工调试或计算机均匀扫描均难以快速完成&局部优化算法可用于7B D C I^质量分析器的电压参数优化&8:3-%;P等(>N)采用梯度下降法得到了>c*i>)"的质量分辨率#g$%1等(>+)% 230-\50-等(())%V$$/等((>)采用'5%Z5.D750ZNMM质谱学报!!第!!卷Copyright©博看网. All Rights Reserved.单纯形算法分别得到了N c)i>)!%*c(i>)"% >c>i>)"的质量分辨率&但以上算法仅能寻找局部最优参数"最终结果受人为设定的初始值影响&7-..0=((()借助全局优化算法"采用遗传算法D'5%Z5.D750Z单纯形混合算法"先在大参数空间下粗略搜索出优势区域"再对其进行精细优化"得到了c)i>)!的质量分辨率&但遗传算法存在参数较多%收敛较慢等问题&相比于上述算法"粒子群!e8I$算法具有搜索速度快%对初始值不敏感%全局搜索能力强%参数易调整等优点&本工作拟采用e8I算法"结合离子光学模拟"以优化7B D C I^质量分析器的电压参数#并通过测试e8I算法和改进粒子群!S e8I$算法的性能"对比(种算法的优化结果&A!实验部分A B A!离子光学模拟采用离子光学模拟软件8S7S I'D()()((*)进行离子飞行轨迹计算&模拟实验忽略了机械结构误差%电源波动%残余气体和空间电荷效应的影响"因此"与实际应用结果存在一定差异& A B A B A!模型!7B D C I^质量分析器模型使用O`O的7B D C I^质量分析器几何结构(>("(!)"由多个圆环电极共轴排列而成"总长度+"!<<"包括(个反射镜和>个漂移管"每个反射镜由N 个电极组成"网格划分精度为)c)"<<'X.4Z"示于图>&A B A B C!初始设置!离子初始状态与文献((")的设置相同&以能量O K>c"]5#"能量标准偏差`K N c"5#的>**26H为研究对象"离子数量设为")个#离子起始位置位于7B D C I^质量分析器的几何中心平面"J K0K)"位置标准偏差J K 0K><<#离子在?方向上的出射角度为)<.0Z"角度标准偏差K K>c"<.0Z#各参数均为高斯分布&A B A B F!质量分辨率计算!7B D C I^质谱仪的工作过程为*离子从离子源出发"到达7B D C I^质量分析器后被捕获"进行多次反射飞行&离子从分析器内某一点出发经(个反射镜反射后回到该点"视为飞行>圈&当飞行一定时间后"离子被引出检测器"完成检测&整个检测过程可视为(部分*>$离子从离子源.穿过/ 7B D C I^质量分析器直达检测器#($离子在7B D C I^质量分析器进行多次反射&7B D C I^质量分析器的质量分辨率((M)计算公式如下*A<W<&<B(&BWB9;Y E B)(!&/Y E&/)$!>$式中"B9;是离子从离子源.穿过/分析器直接到达检测器的飞行时间#E是飞行圈数#B)是离子在7B D C I^质量分析器内飞行>圈的时间# &/是离子从离子源.穿过/分析器直接到达检测器的时间展宽#&/)是7B D C I^质量分析器飞行>圈引起的时间展宽&当E无限大时"B9;的影响被抵消"质量分辨率趋于B)'!(&/)$"此时得到的质量分辨率称为极限分辨率A<"<0Q"其受限于7B D C I^质量分析器引起的时间展宽&本工作采用.时间焦点/固定的聚焦模式((?)"计算离子飞行半圈后回到起点所在平面的C I^标准偏差 ;$1"从而计算得出A<"<0Q&此聚焦模式将7B D C I^质量分析器的(个反射镜视为完全相同的单元"采用离子飞行半圈方式得到的极限质量分辨率评价电压参数的优劣"不仅可以节约模拟时间"还可以减少优化变量"提高优化效率&图A!>131P G中的32=:P I质量分析器模型I#9J A!32=:P I4&))&%&'(N0-4"!0'#%>131P G+ M M第"期!!黄!奇等*基于粒子群算法的多次反射飞行时间质量分析器电压优化Copyright©博看网. All Rights Reserved.A B C !粒子群算法e 8I 算法是一种基于群体协作的搜索算法((N )"按照一定的规则更新随机初始化的粒子速度和位置"在多维搜索空间中搜索出最优解"实现最优的目标适应值&在7B D C I ^质量分析器电压参数优化中"粒子的位置代表>组电压参数"目标适应值是7B D C I ^质量分析器所能实现的极限质量分辨率A <"<0Q &e8I 算法的作用是通过指导电压参数的更新"寻找使7B D C I ^质量分析器的A <"<0Q 达到最大的电压参数&假设待优化问题的搜索空间为@维"第*个粒子的位置表示为U *K !J *>"J *("+"J *N $"其所经历的最优位置表示为:*K !;*>";*("+";*N $"即:R 56;"整个群体所经历的最优位置表示为:&K !;&>";&("+";&N $"即:W R 56;&粒子*的移动速度表示为T *K !+*>"+*("+"+*N $"根据以下公式进行粒子速度和位置的更新*T /Y>*W S T /*Y '>->!:R 56;X U /*$Y '(-(!:W R 56;X U /*$!($U /Y >*W U /*Y T /Y >*!*$式中"群体中粒子个数*K>"("+"3#'>和'(为学习因子#->%-(为)!>之间的随机数#/为当前迭代次数#S 为速度权重&A B F !改进粒子群算法速度权重S 控制着粒子的上一步速度对当前速度的影响程度"较大的S 有利于增强算法的全局搜索能力"较小的S 有利于增强算法的局部搜索能力&e 8I 算法中的S 为固定值"无法同时实现较强的全局和局部搜索能力&本文采用速度权重线性衰减的改进e 8I!S e 8I $算法((+)"对S 的取值方式进行改进"根据式!!$"使S 随着迭代的进行从最大值线性减小至最小值"以兼顾全局和局部搜索能力&S W S <0Q X !S <0Q X S <4/$Z */,-'*/,-<0Q !!$式中"S <0Q 为速度权重最大值"S <4/为速度权重最小值"*/,-为当前迭代次数"*/,-<0Q 为最大迭代次数&C !结果与讨论C B A !T >P 算法不同初始参数性能测试比较e 8I 算法不同初始参数的优化结果"设定*/,-<0Q K !))"S K )c N "'>K '(K >c "&为避免偶然性"各条件下算法重复运行()次"得到统计结果&C B A B A !种群规模的影响!本实验测试了种群规模分别为*)和M )个粒子数的算法优化效果"列于表>&可以看出"不同种群规模下的优化结果相差不大"且随着种群规模增加>倍"优化时间成本也增加>倍&为降低优化时间成本"采用粒子数为*)的种群规模&表A !不同种群规模的优化结果对比:&;'0A !+"47&-#)"%"*"7/#4#N &/#"%-0)K '/)@#/,!#**0-0%/7"7K '&/#"%)#N 0)种群规模e $9-%0;4$/64P 5平均值750/\0%-5最大值70Q 4<-<\0%-5标准偏差8;0/Z 0.Z Z 5\40;4$/*)!)))))M )M ?M N >)))))M )*N ))))M >*!N ">"))))C B A B C !速度权重的影响!本实验测试了S 分别为)c N %)c "和)c (时的e 8I 算法优化效果"列于表(&可以看出"S K )c N 优化结果的平均值和最大值均优于另外(种情况"但收敛性较差"收敛成功率仅为>)b #S K )c (优化结果的平均值最小"但收敛性最好"收敛成功率达+"b &S 的测试为S e 8I 算法的参数设置提供了参考&表C !不同速度权重的优化结果对比:&;'0C !+"47&-#)"%"*"7/#4#N &/#"%-0)K '/)@#/,!#**0-0%/4速度权重S平均值750/\0%-5最大值70Q 4<-<\0%-5标准偏差8;0/Z 0.ZZ 5\40;4$/收敛成功率8-::566.0;5$1:$/\5.X5/:5'b )c N !)))))M )M ?M N >)))))>))c "**))))">))))N ))))M ))c (>!))))"N ))))>?))))+")?M 质谱学报!!第!!卷Copyright ©博看网. All Rights Reserved.C B C !T >P 算法与1T >P 算法性能对比将e 8I 算法与S e 8I 算法进行对比"*/,-<0Q K !))"'>K '(K >c ""优化结果列于表*&e 8I 算法的S K )c N "S e 8I 算法的S <0Q K)c N %S <4/K)c (&S e 8I 算法优化的最优A <"<0Q K N c >i >)""相比e 8I 算法的M c )i >)"提高了**b "平均优化结果提高了*"b "标准偏差降低了(+b &S e 8I 算法优化的最优电压参数7>!7N 分别为()+"c N (%+>)c +)%>+>N c >?%N M c +N %(>)c +M %J !>(N c +!%J M "M >c ++%J *>N (c M "#&可见"S e 8I算法具有更好的优化结果以及更高的稳定性&表F !T >P 算法和1T >P 算法的优化结果:&;'0F !P 7/#4#N &/#"%-0)K '/)"*T >P&'9"-#/,4&%!1T >P&'9"-#/,4算法A %X $.4;3<平均值750/\0%-5最大值70Q 4<-<\0%-5标准偏差8;0/Z 0.Z Z 5\40;4$/e 8I !)))))M )M ?M N >)))))S e 8I"!))))N )+>!)N ))))图C !1T >P 算法和T >P 算法最优结果中当代最优!4 4&6随迭代次数的变化情况I #9J C !W &-#&/#"%"*."%/047"-&-("7/#4&'!4 4&6@#/,/,0%K 4;0-"*#/0-&/#"%)#%/,0"7/#4&'-0)K '/)"*1T >P&'9"-#/,4&%!T >P&'9"-#/,4S e 8I 算法和e 8I 算法当代最优A <"<0Q 随迭代次数的变化情况示于图(&e 8I 算法能够寻找到多个局部最优解"在M )次迭代时"得到超过!i >)"的分辨率"而S e 8I 算法需要())次迭代才能得到同等结果&e 8I 算法从M )代到!))代优化中"迭代次数增加了约M 倍"适应值仅提高了")b "在优化后期"e 8I 算法出现了在最优解附近振荡的现象&S e 8I 算法在前>")代"A <"<0Q 处于较低水平">")!(()代之间"A <"<0Q 出现大幅增长"随后算法进入收敛状态"适应值趋于稳定&以上结果表明"e 8I 算法具有较强的前期搜索能力"但难以保持稳定"无法进行精细化搜索"而S e 8I 算法拥有较强的精细化搜索能力"一旦搜索到较优区域"能够很快搜索到当前区域的最优值"并且保持稳定#e 8I 算法迭代!))次仍无法收敛"而S e 8I 算法在(")代左右即完成收敛"优化效率高于e 8I 算法&e 8I 算法和S e 8I 算法优化的N 个电压更新步长随迭代次数的变化情况示于图*&e 8I 算法的迭代步长虽然在粒子群算法本身的特性下逐渐衰减"但在优化后期"电压更新步长依然处于比较宽的范围内&电压在较宽范围内的变化将造成A <"<0Q 的显著变化"这是e 8I 算法无法收敛%在最优值附近振荡的原因&而S e 8I 算法的速度权重线性衰减策略能够很好地控制电压更新步长的减小"在前期与e 8I 算法的电压更新步长范围基本相同"随着迭代次数的增加"电压更新步长降低"进而实现对优势区域的局部精细化搜索"在())!*))代之间"大多数电极电压已完成收敛"保持稳定&因此"S e 8I 算法具有较好的收敛性和较快的收敛速度"且对7B D C I ^质量分析器电压优化问题具有更好的适用性&e 8I 算法采用的速度权重是固定的"无法控制电压更新步长收缩至优化问题局部精细化搜索所需的步长范围&而S e 8I 算法改进了速度权重的取值方式"采用线性衰减的方式使速度权重随迭代次数逐渐减小"从而使电压更新步长逐渐收缩至优化问题局部精细化搜索所需的步长范围&这种改进使S e 8I 算法具备了良好的全局和局部搜索效果"能够获得更好的优化结果&C B F !1T >P 算法最优结果分析7B D C I ^质量分析器的电压影响着飞行时间相对偏差与能量相对偏差的关系"飞行时间相对偏差对能量相对偏差的容忍度越高"则质量分辨率越高&在最优电压参数下"半圈飞>?M 第"期!!黄!奇等*基于粒子群算法的多次反射飞行时间质量分析器电压优化Copyright ©博看网. All Rights Reserved.注*图中右下角数字代表电极编号图F !C 种算法优化的电压更新步长随迭代次数的变化情况I #9J F !W &-#&/#"%"*/,08"'/&90K 7!&/0)/07)#N 0@#/,/,0%K 4;0-"*#/0-&/#"%)*"-/,0/@"&'9"-#/,4)"7/#4#N &/#"%图O !离子半圈飞行时间相对偏差随能量相对偏差的变化关系I #9J O !20'&/#80!08#&/#"%"*#"%,&'*=/K -%/#40"**'#9,/&)&*K %./#"%"*-0'&/#80!08#&/#"%"*0%0-9(行时间相对偏差与能量相对偏差的关系示于图!&通过对半圈飞行时间相对偏差与能量相对偏差变化关系进行"次多项式拟合"得到*&B B )W (c >Z >)X >)X (c +Z >)X"Z &O O )Y>c ""Z >)X*!Z &O O $)(Y )c (!!Z `O $)*Y(c )!!Z &O O $)!X )c ??!Z &O O $)"!"$对&O 'O )求导后">阶和(阶项系数接近于)"可认为优化后的电势分布实现了时间关于能量的(阶聚焦&在能量分散为!>"))L ("$5#离子状态下"半圈飞行时间最大相对偏差在>c *i>)J M 以内&对能量分散更大的!>"))L ")$5#离子满足同样的(阶聚焦"半圈飞行时间最大相对偏差在>i >)J "以内&当离子能量大于初始平均能量!>"))5#$时"时间偏差随能量偏差增长较快#而离子能量小于初始平均能量!>"))5#$时"时间偏差随能量偏差增长较慢&这表明"相比能量更小的离子"优化后的静电场难以对能量更大的离子进行能量补偿&因为能量更大的离子路径更长"经历了还未优化后的静电场"表现出较差的能量补偿#而能量更小的离子经历的静电场均是经过优化的"能够较好地保证能量补偿&这表明算法只针对离子实际经历的静电场进行优化"而对路径外的静电场优化力度不足"优化结果与所使用的离子能量状态有关&上文计算A <"<0Q 的方式是离子飞行半圈"&/被忽略&为了与实际应用结合"测试了&/K ()/6时"质量分辨率与飞行圈数的关系&将7B D C I ^质量分析器的(个反射镜对称施加相同电压"根据离子每次飞行至7B D C I ^质量分析器中心平面"测量其飞行时间和飞行时间标准偏差"从而计算出当前飞行圈数下的质量分辨率"示于图"&可见"质量分辨率随飞行圈数的增加而增加"最终接近A <"<0Q &但实际应用中"离子的飞行圈数并不能任意增加"主要受限于以下几个方面*>$真空度的限制&分析器(?M 质谱学报!!第!!卷Copyright ©博看网. All Rights Reserved.内部的残余气体造成离子损失"进而降低信号强度#($电源稳定性的限制&长时间的飞行对电源稳定性提出了更高要求#*$空间电荷效应的影响&由于离子间存在空间电荷效应"在长时间的相互作用下"将影响离子飞行轨迹的稳定#!$检测时间短的需要&当离子飞行()<6!M ))圈$时"能够达到"i>)"的质量分辨率&7B D C I ^质量分析器的多圈飞行结果为仪器调试提供了重要参考&图Q !质量分辨率随离子飞行圈数的变化情况I #9J Q !W &-#&/#"%"*4&))-0)"'K /#"%@#/,/,0%K 4;0-"*#"%*'#9,//K -%)F !结论本研究开发了基于e 8I 算法的7B D C I ^质量分析器电压参数优化方法"并对e 8I 算法进行改进"此方法可以优化7B D C I ^质量分析器的电压参数"以获得更高的质量分辨率&基于8S 7S I '模拟"采用节省时间%减少优化变量的离子飞行半圈计算A <"<0Q 的方式&对O K>c "]5#% `KN c "5#% Q K =K><<% )K *K>c "<.0Z 的>**26H"&/K )/6时"e 8I 算法获得A <"<0Q 超过M )))))"S e 8I 算法获得A <"<0Q 为Nc >i>)"#&/K()/6时!飞行()<6%M ))圈$"S e 8I 算法获得质量分辨率为"i >)"&S e 8I 算法优化得到的最佳结果能够实现7B DC I ^质量分析器的时间关于能量的(阶聚焦"离子的半圈飞行时间偏差在>c *i>)J M以内&e 8I 算法的平均结果为!)))))"最优结果为M )M?M N "标准偏差为>)))))#S e 8I 算法的平均结果为"!))))"最优结果为N )+>!)"标准偏差为N ))))&S e 8I 算法相比于e 8I 算法优化的平均结果提高了*"b "最优结果提高了**b "标准偏差降低了(+b "具有更好的优化结果和稳定性&S e 8I 算法的速度权重线性衰减策略有效地控制了电压更新步长"能够满足7B D C I ^质量分析器电压参数优化问题全局搜索和局部搜索的要求"具有较好的收敛性和收敛速度&本研究表明"S e 8I 算法能够解决7B D C I ^质量分析器电压参数的优化问题"具有较好的适用性和稳定性"可以提高电压参数的优化效率&参考文献(>)!BA @S I 'I #A A "^S E S e e I #S "@`B B S 2Oe,&S /9-.6-4;$1.56$%-;4$/4/;4<5D $1D 1%4X 3;<066695:;.$<5;.=*0346;$.4:0%95.695:;4\5*4/9-.6-4;$1.56$%-;4$/4/;$1<066695:;.$<5;.=(,)&7066895:;.$<5;.=B 5\45[6"()>M "*"!M $*?*N D ?"?&(()!a I `8E T&C 4<5D $1D 1%4X 3;<066695:;.$<5;.=*4/;.$Z -:;4$/;$;35R 064:6*;4<5D $1D 1%4X 3;<066695:;.$<5;.=(,)&7066895:;.$<5;.=B 5\45[6"()>?"*M !>$*N M D >)+&(*)!黄建鹏"贺玖明"朱辉"李铁钢"黄正旭"莫婷"李梅&国产高分辨飞行时间质谱仪在药物分子结构鉴定中的应用(,)&质谱学报"()>M "*?!"$*!*>D !*+> A 'W,40/95/X "G `,4-<4/X "Y GT G -4"E S C 45X 0/X "GT A 'WY 35/X Q -"7IC 4/X "E S754&A /0%=646$1930.<0:5-;4:0%<$%5:-%561$.6;.-:D ;-.54Z 5/;414:0;4$/R =Z $<56;4:34X 3.56$%-;4$/;4<5D $1D 1%4X 3;<066695:;.$<5;5.(,)&,$-./0%$1234/5657066895:;.$<5;.=8$:45;="()>M "*?!"$*!*>D !*+!4/234/565$&(!)!解迎双"张欢"王娟"王波&实时直接分析D 串联质谱法快速测定环境水体中涕灭威及其代谢物(,)&质谱学报"()(("!*!>$*++D >)N &U S `V 4/X 63-0/X "Y GA 'W G -0/"gA 'W,-0/"gA 'W a $&B 094ZZ 5;5.<4/0;4$/$10%Z 4:0.R0/Z 4;6<5;0R $%4;564/[0;5.5/\4.$/<5/;R =.50%D ;4<5Z 4.5:;0/0%=646D ;0/Z 5<<066695:;.$<5;.=(,)&,$-./0%$1234/5657066895:;.$<5;.=8$:45;="()(("!*!>$*++D >)N !4/234/565$&(")!林黛琴"万承波"邱萍"刘花梅&液相色谱D 串联质谱法快速测定食品中!种黄色工业染料(,)&质谱学报"()>*"*!!*$*>?)D >?N &E S '@04_4/"gA '235/X R $"F S T e 4/X "E S TG -0<54&B 094ZZ 5;5.<4/0;4$/$11$-.34X3=5%%$[*?M 第"期!!黄!奇等*基于粒子群算法的多次反射飞行时间质量分析器电压优化Copyright ©博看网. All Rights Reserved.Z=564/1$$Z6R=G e E2D78'78(,)&,$-./0%$1234/5657066895:;.$<5;.=8$:45;="()>*"*!!*$*>?)D>?N!4/234/565$&(M)!gI E E'S O G"2A8A B`8A&A/5/5.X=D46$:3D .$/$-6<-%;4D9066;4<5D$1D1%4X3;<066695:;.$<5D ;5.:$/646;4/X$1;[$:$0Q40%5%5:;.$6;0;4:<4..$.6(,)&S/;5./0;4$/0%,$-./0%$17066895:;.$<5;.="())*"((?!($*(>?D(((&()!e E A y g B"@S2O`E C"82G`S@`'a`B W`B 2&7-%;49%5D.51%5:;4$/;4<5D$1D1%4X3;<066695:D ;.$<5;.=(,)&S/;5./0;4$/0%,$-./0%$17066895:D ;.$<5;.="()>*!*!+'*")$*>*!D>!!&(N)!@S2O`E C"e E A y g B"a`2O`B A"2Y I O T"W`S88`E G"GA`C C'`B`",`82G2"O S'8`E g"e`C B S2O7"82G`S@`'a`B W D`B2"8S7I'A"V A#I B7S&A34X3D95.D 1$.<0/:5<-%;49%5D.51%5:;4$/;4<5D$1D1%4X3;<066695:;.$<5;5.0/Z46$R0.6590.0;$.1$.;35.5650.:3[4;35Q$;4:/-:%54(,)&'-:%50.S/6;.-D<5/;60/Z75;3$Z64/e3=64:6B5650.:385:;4$/A*A::5%5.0;$.6"895:;.$<5;5.6"@5;5:;$.60/ZA66$:40;5Z`_-49<5/;"()>""???*>?(D>N N& (+)!gI E^B'"g S`'G I E C Y^"A C A'A8I#@"a`2O@"a E A T7O"a I B W7A''2G"G`B^T B C G^"O IgA E8O A7"O B`S78"E S C#S'I#V A"E T''`V@"7A'`A#"'`S@G`B B@"B I8`'a T82G7"82Gg D`S O G A B@E"8C A',A,"Y T a`B O&S8I E DC B A e36<-%;4D.51%5:;4$/;4<5D$1D1%4X3;<0666590D.0;$.'695:;.$<5;5.(,)&S/;5./0;4$/0%,$-./0%$17066895:;.$<5;.="()>*!*!+'*")$*>(*D>**& (>))g S`'G I E C Y^"a`2O@"a E A T7O"a I B W D 7A''2G"a B`S C`'^`E@C7"2A O S B E SBa"W`I B W`8"G`B^T B C G^"G I E C,@"O IgA E8O A7"O B`S78"E T''`V@"7A'`A#"7`'v'@`Y,"'`S@G`B B@"B I8`'a T82G7"82Gg`S O GA B@E"82G Dg`'O A"8S7I'S8,"8C A',A,"gI E^B'"Y T a`BO&706656$15Q$;4::0%:4-<46$;$95694/Z$[//-:%50.1$.:56(,)&'0;-.5"()>*"!+N!!"!$**!M D*!+&(>>)S C I V"82GT B V e"gA@A7"'A S7S8"8I'I@A C"7S C A G"A B A S^"C A O A7S'`A"I O A@A O"I Y AgA A"gI E E'S O G& 84/X%5D.515.5/:534X3D9.5:464$/<066<506-.5D<5/;[4;30<-%;4.51%5:;4$/;4<5D$1D1%4X3;<066695:;.$X.093(,)&e3=64:0%B5\45[2"()>*"N N!>$*)>>*)M&(>()82GT B V e"gA@A7"S C I V"A B A S^"'A S7S8"8I'I@A C"gI E E'S O G"8G2G`e T'I##A"87I B B A2"V T A'2&A34X3D.56$%-;4$/<-%;4D.51%5:;4$/;4<5D$1D1%4X3;<066695:;.$X.0931$.9.5:464$/<066<506-.5D<5/;60;B S O`''8E IgB S(,)&'-:%50.S/6;.-D<5/;60/Z75;3$Z64/e3=64:6B5650.:385:;4$/a*a50<S/;5.0:;4$/6[4;370;5.40%60/ZA;$<6"()>!"**"**+D"*&(>*)O T E S O I#S"A E W I B A A"A C A'A8I#@"A82G`Be"a E A T7O"2A O S B E SBa"G`B DE`B C A"GT A'W g,"O A B C G`S',"E S C#S'I#V A"E T''`V@"7A'`A#"7I T W`I C7"82Gg`S O GA B@E"g`E O D`B A"g S`'G I E C Y^&706656$163$.;D%4\5Z!+8:"")8:"?)A6"?*a.0/Z6;0R%5>+M G X/-:%4Z56(,)&'-:%50.e3=64:6A"()()">))(*>(>++)& (>!)e S`2GA2Y`O A"8G2G`e T'I##"2A B C D `B G O"a A C2G`E@`B,2"Y W A',A B`^"E S@@S2O8'"gI E E'S O G"GT V"W B S^D^S C G aI&@5\5%$9<5/;$1034X3.56$%-;4$/46$DR0.6590.0;$.1$.6;-Z=$15Q$;4:Z5:0=6(,)&'-D:%50.S/6;.-<5/;60/Z75;3$Z64/e3=64:6B5D650.:385:;4$/a*a50<S/;5.0:;4$/6[4;370;5.4D0%60/ZA;$<6"())N"(M M!>+'()$*!">)D!">!& (>")e E A y g B"@S2O`E C"2Y I O T"W`S88`E G"e`C B S2O7"B`S'G`S7`B O"82G`S D@`'a`B W`B2"V A#I B7S&S6$R0.6590.0D ;4$/R=;4<5D$1D1%4X3;<066695:;.$<5;.=1$.%$[D5/5.X=.0Z4$0:;4\54$/R50<10:4%4;456(,)&'-:%50.S/6;.-<5/;60/Z75;3$Z64/e3=64:6B5650.:385:;4$/a*a50<S/;5.0:;4$/6[4;370;5.40%60/ZA;$<6"())N"(M M!>+'()$*!"M)D!"M!&(>M)a`''`B g G&A X0;5Z5%5:;.$6;0;4:4$/;.09;$ .595;4;4$-6%=<506-.5;35:30.X50/Z<'?$1%0.X55%5:;.$69.0=4$/6(,)&A/0%=;4:0%235<46;.=">++?"M+!()$*!>M(D!>M N&(>?)@A GA'7"^S8G7A'B"G`a`B I"B A e De A e I B C7"A E C8C`S''"Y A,^7A'@"\0/Z5.Y A'@`g,&A/5[;=95$15%5:;.$6;0;4: 4$/;.091$.6;$.0X5$1106;4$/R50<6(,)&B5\45[$18:45/;414:S/6;.-<5/;6">++N"M+!>$*?M D N*& (>N)82GT E C Ya`"O`E E V,7"'S2I E I^^2"E I'W,"B V A'8"a B I@`T B7&2$/6;.-:D;4$/0/Z64<-%0;4$/$10<-%;4D.51%5:;4$/;4<5D$1D1%4X3;<066695:;.$<5;5.0;;35-/4\5.64;=$1!?M质谱学报!!第!!卷Copyright©博看网. All Rights Reserved./$;.5Z0<5(,)&'-:%50.S/6;.-<5/;60/Z75;3D$Z64/e3=64:6B5650.:385:;4$/a*a50<S/;5.0:D ;4$/6[4;370;5.40%60/ZA;$<6"()>M"*?M*(">D(""&(>+)g I E^B'"`B S C C7"7A B U W"82Gg D `S O G A B@E&A<-%;4D.51%5:;4$/;4<5D$1D1%4X3;<0666590.0;$.1$.46$R0.4:9-.414:0;4$/$1.0Z4$D0:;4\54$/R50<6(,)&G=95.14/5S/;5.0:;4$/6"()>>">++!>'*$*>>"D>((&(())2G A T#`A Te"@`E A G A V`e"Z5^B A'2`W"5%A a S B8"E I B V,"7`B B`B V"B I8`'Da T82G7"82Gg`S O GA B@E"gI E^B'&e S E W B S7"0<-%;4D.51%5:;4$/;4<5D$1D1%4X3;<066695:;.$<5;5.1$.694.0%(D8*0;W A'S E(,)&'-D:%50.S/6;.-<5/;60/Z75;3$Z64/e3=64:6B5D650.:385:;4$/a*a50<S/;5.0:;4$/6[4;370;5.4D0%60/ZA;$<6"()>M"*?M*(>>D(>"&((>)V I I',g"e A B O V G"S7O a"O S7W@"O S7V O&@564X/6;-Z=1$.0<-%;4D.51%5:;4$/;4<5D$1D1%4X3;<066695:;.$X.0931$.\5.=63$.;%4\5Z/-:%54(2)&e.$:55Z4/X6$1;352$/15.5/:5$/A Z\0/:564/B0Z4$0:;4\5S6$;$958:45/:5!A B S8()>!$"C$]=$",090/*,$-./0%$1;35e3=64:0%8$:45;=$1,090/"()>"&((()7T B B A V O&C35Z564X/0/Z$9;4<4P0;4$/$10 <-%;4D.51%5:;4$/;4<5D$1D1%4X3;<066D695:;.$<5;5.1$.a0.4-<;0X X4/X[4;3/`U I0/Z$9;4<4P0;4$/$1;35>*?U5\5;$[4;3`U I D())(@)&20/0Z0*7:W4%%T/4\5.64;="()>N&((*)@A G E@A&84<4$/1$.;3595.6$/0%:$<9-;5.4/ .51%5:;4$/(,)&S/;5./0;4$/0%,$-./0%$17066895:;.$<5;.="()))"())!>'*$**D("&((!)82GT B V e"I O A@A O"8G2G`e T'I#8"8I'I@A C"C A O A7S'`A"gA@A7"gI E E'S O G"V A7A Y A O SV&7-%;4D.51%5:;4$/;4<5D$1D1%4X3;<066695:;.$X.0931$.63$.;D%4\5Z.0Z4$0:;4\54$/6(,)&C35`-.$950/e3=64:0%,$-.D/0%A"())+"!(!*$**!*&((")王永生"田玉林"王均英"周小红"黄文学&用于兰州彭宁离子阱的多反射飞行时间质量分析器的设计和优化(,)&原子核物理评论"()>?"*!!*$* M(!D M(+&gA'W V$/X635/X"C S A'V-%4/"gA'W,-/=D 4/X"Y G I T U40$3$/X"GT A'W g5/Q-5&@564X/0/Z$9;4<4P0;4$/$10<-%;4D.51%5:;4$/;4<5D$1D1%4X3;<066695:;.$<5;5.1$.E e C(,)&'-:%50.e3=64:6B5\45["()>?"*!!*$*M(!D M(+!4/234/565$&((M)V A#I B7"#`B`'C2G S O I#A"GA8S',"O I Y E I#a"W A#B S O7"C B T^A'I#A& e%0/0.<-%;4D.51%5:;4/X;4<5D$1D1%4X3;<0660/0%=DP5.[4;30h4X D60[4$/90;3(,)&e3=64:6e.$:5Z40"())N">**+>D!))&((?)@S2O`EC"V A#I B7S"E A'W,"e E A y g B"E S e e`B C g"W`S88`E G"82G`S@`'Da`B W`B2&@=/0<4:0%;4<51$:-66341;4/<-%;4D9%5D.51%5:;4$/;4<5D$1D1%4X3;<066695:;.$<5;5.6(,)&S/;5./0;4$/0%,$-./0%$17066895:;.$<5;.="()>?"!>(*>D?&((N)O`''`@V,"`a`B GA B C B&e0.;4:%56[0.< $9;4<4P0;4$/(2)&e.$:55Z4/X6$1S2''f+"D4/;5.D/0;4$/0%:$/15.5/:5$//5-.0%/5;[$.]6"S```">++"*>+!(D>+!N&((+)8G SV"`a`B GA B C B2&`<94.4:0%6;-Z=$190.;4:%56[0.<$9;4<4P0;4$/(2)&e.$:55Z4/X6$1;35>+++2$/X.566$/`\$%-;4$/0.=2$<9-;0;4$/D2`2++!20;&'$&++C G N!)M$"S```">+++*>+!"D>+")&!收稿日期*()(*D)*D>!#修回日期*()(*D)"D)!$"?M第"期!!黄!奇等*基于粒子群算法的多次反射飞行时间质量分析器电压优化Copyright©博看网. All Rights Reserved.。
Experimental investigation on capillary force of composite wick structure
Experimental investigation on capillary force of composite wick structure by IR thermal imaging cameraYong Tang,Daxiang Deng *,Longsheng Lu,Minqiang Pan,Qinghui WangSchool of Mechanical and Automotive Engineering,South China University of Technology,Wushan Road,Guangzhou 510640,Chinaa r t i c l e i n f o Article history:Received 1January 2009Received in revised form 3October 2009Accepted 19October 2009Keywords:Composite wick Meniscus Heat pipes Capillary forceIR thermal imaginga b s t r a c tA novel sintered–grooved composite wick structures has been developed for two-phase heat transfer devices.With ethanol as the working fluid,risen meniscus test is conducted to study the capillary force of wick structures.Infrared (IR)thermal imaging is used to identify and locate the liquid meniscus.The effects of sintered layer,V-grooves and powder size on capillary force are explored.The results show that the capillary force of composite wick structures is larger than that of grooved and sintered ones.Interac-tion wetting between groove and sintered powder happens during the liquid rise in composite wick,which provides an additional source of capillary force.It exhibits a variation of capillary force of compos-ite wicks with different powder size due to the difference of open pore size and quantity in sintered por-ous matrix.Ó2009Elsevier Inc.All rights reserved.1.IntroductionDevelopment of heat pipes,vapor chambers and capillary pumped loops (CPL)is motivated by the thermal management of electronic components.These heat transfer devices work via two-phase flow.By evaporation and condensation,heat is transported from one end to the other or spread to a large area.Two-phase heat transfer devices consist of a number of components,such as evap-orator,wick and condenser.The wick shows great effects on the heat transfer performances of these devices.Wick structures are generally grooved or sintered.Sintered wicks have large capillary force but low permeability,while grooved wicks have high perme-ability but small capillary force [1].Capillary force and permeabil-ity are somehow contradictory in a single wick.However,composite wicks can balance these two competing factors and im-prove heat transfer performances.Currently,researches have been focused on composite wick structures [2–4].Hwang et al.[2]ob-tained modulated composite wicks by making grooves over a thin uniform porous layer.The modulated composite wick improved heat transfer performance by providing extra cross-sectional area for enhanced axial capillary liquid flow and extra evaporation sur-face area.Wang and Catton [3]investigated a composite structure with a thin porous layer on the triangular groove.It was found that evaporation heat transfer in the composite structure was three to six times higher than that in the groove without a porous layer.Capillary force of wick plays the driving force of the circulation of working liquid in two-phase heat transfer rge capil-lary force makes these devices work well.Capillary force of wick has been determined mainly through two methods.One is the bub-ble point test [5].Pressurized gas is applied to one end of a wick saturated with liquid.With the pressure gradually increasing,the point that bubbles appear on the opposite end of the wick is re-corded for the capillary force.The other method is the risen menis-cus test [6,7].One end of a wick is dipped into the working liquid.Then the liquid level rises in the wick sample until the pressure on both sides of the meniscus equilibrates.Nowadays,the latter method has been widely used due to its convenience.Chen et al.[8]observed the liquid fronts in the inclined wick by optical micro-scope (OM)and CCD camera.Holley and Faghri [9]obtained the closed form solution of effective pore radius and permeability of wicks by sight and weight change.Nevertheless,visualization of the meniscus by sight may be not accurate because most of the working liquid is colorless and transparent.In the case of micro grooved or sintered wick structures,it is difficult to locate the meniscus.This problem may be solved with the technique of infra-red (IR)thermal imaging.IR thermal imaging can provide accurate and non-contact measurement of temperature field [10].It has been used for the test of thermal performance and heated flow of liquid film of two-phase heat transfer devices [11–15].Due to the infrared emissivity difference between wick samples and working liquid,the meniscus can be accurately located by IR ther-mal imaging.Thus,the rising of the wetted height driven by capil-lary force can be investigated.0894-1777/$-see front matter Ó2009Elsevier Inc.All rights reserved.doi:10.1016/j.expthermflusci.2009.10.016*Corresponding author.Tel./fax:+862087114634.E-mail address:dengdaxiang88@ (D.Deng).Experimental Thermal and Fluid Science 34(2010)190–196Contents lists available at ScienceDirectExperimental Thermal and Fluid Sciencejournal homepage:www.else v i e r.c o m /l o c a t e /e t fsThe present study proposes a novel sintered–grooved compos-ite wick structure by covering a layer of sintered copper powderon micro V-grooves.With the risen meniscus test method,the cap-illary force of the wick is investigated by the visualization of IR thermal imaging camera.2.Experimental2.1.Fabrication of composite wick structureSintered–grooved composite wick structure is shown in Fig.1.Micro V-grooves were made in copper base with the thickness of 1.0mm.Then copper powders were sintered on the grooves.Dur-ing sintering,adhesion and mass transfer happened by the driving of surface tension.Different powder particles bonded together.Due to the same material of copper,powder particles and groove fins also partially bonded together,as seen in Fig.2.Thus,sintered–grooved composite wick formed.The copper powders (supplied by ACuPowder International,LLC,USA)are of irregular morphol-ogy,with purity of 99.3%,produced by water atomization method [16].To optimize the sintered powder size,four composite wick samples were made.A grooved sample without powder and a sin-tered sample without grooves were also studied for comparison.For the sintered sample,a 0.5mm thick powder layer was sintered on 1.0mm thick copper base .All the samples had the same dimen-sions,with length 100mm and width 10mm,as shown in Table 1.Copper powder layer was sintered at temperatures 950±10°C in a tube furnace for half an hour under a reduction stream of 10%hydrogen and 90%argon.2.2.Experimental apparatusAn apparatus of the risen meniscus test is set up to determine the capillary force of wick structures,as shown in Fig.3.It consists of vertically adjusting device,sample fixing device,reservoir,IR thermal imaging camera,glass cover with a hole for IR thermal imaging visualization,PLC controller and PC.The classical working liquid,ethanol,is used in the experiment.At first,test sample is fixed in vertical in the sample fixing device.Secondly,the sample moves down to dip into ethanol by the vertically adjusting device in a constant speed.The dipped length of samples in the liquid is 2mm.At last,IR camera records the meniscus rising process in 1min from the point that the sample just dips into ethanol.Ambi-ent temperature is 30°C.The glass cover is used to ensure the cir-cumstance airflow stable.The volatilization of ethanol is in a constant speed.Therefore,the difference between samples can be observed accurately.2.3.Meniscus locating method in IR thermal imagingA FLIR ThermaCAM SC3000IR camera was used with a thermal sensitivity of 0.02K at 30°C and an accuracy of 1%for tempera-tures below 150°C of the full scale.The back-side of the samples was painted black to provide a uniform emissivity.Due to the infrared emissivity difference between copper and ethanol,differ-ent temperature distributions between samples and ethanol were displayed in IR thermal images.The meniscus could be accurately located as follows.As shown in Fig.4,a measuring line was drawn along the sample from the fixing device to the dipping end.It stoodNomenclature D powder size,l mg gravity acceleration,m/s 2h capillary wetted height,mm h g depth of grooves,mmh t height of total grooves,mm K permeability,mm 2M mass of powder,gr eff effective pore radius,mms spacing interval of grooves,mm T thickness of wick,mm wwidth of grooves,mmGreek symbols4P capcapillary force,Pa 4P cap ,com capillary force of composite wicks,Pa4P cap ,sin capillary force provided by sintered porous zone,Pa 4P cap ,gro capillary force provided by groove bottom zone,Pa 4P cap ,int capillary force provided interaction wetting zone,Pa e porosity,dimensionless l dynamic viscosity,Pa s h liquid–solid contact angle,rad q density,kg/m 3r surface tension,N/mFig.1.Schematic of sintered–grooved composite wickstructure.Fig.2.Sintered bonding of composite wick structure.Y.Tang et al./Experimental Thermal and Fluid Science 34(2010)190–196191for the sample and its length had been measured before the sample dipped into ethanol.A locating point was added in the measuring line.As this point moved vertically along the measuring line,the vertical line moved along the temperature distribution curve until arriving at the inflection point.The position of this inflection point stood for the meniscus.The height result of the point could be cal-culated.Thus,the wetted height,h ,was obtained.Besides,the wet-ted height over time during the whole visualization process could be accurately obtained.Thus,the rising velocity of wetted height could be also studied.3.Analysis of liquid flow characteristic in composite wick structureDuring the risen meniscus test,working liquid rises in a wick.The following assumptions are given:(i)steady-state laminar flow in the wick,(ii)uniform saturation with liquid along the wetted length and (iii)neglecting inertial effects and evaporation of liquid.In the initial time after the composite wick dips into ethanol,the liquid rises along two channels,micro groove and sintered por-ous layer.In the groove channel,liquid flows along a straight line on the groove surface,whereas in the sintered porous channel,li-quid flows tortuously by the open pores in sintered porous layer.The friction resistance of liquid flow in groove is smaller than that in sintered porous layer.Therefore,the liquid rises faster in the groove channel than in the sintered porous layer.However,there are small voids which interconnect the groove and sintered porous layer (Fig.2).The liquid in the groove channel and sintered porous layer does not flow separately.There is an interaction effect be-tween these two rising channels,that is,the fast liquid in groove channel drags the slow liquid in sintered porous layer.By this interaction effect,the liquid in these two channels converges to-gether and a balanced rising velocity reaches between groove and sintered porous layer.We name this kind of flow in composite wick as interaction wetting.It is somewhat similar to the plate-particle interaction during water saturation in the channels with particles in contact with a plate of the same material,which was analyzed by Lechman [17].Due to the interaction wetting,the composite wick is completely saturated except the closed poresTable 1Samples code and specifications.Sample code Powder size,D mm Mass ofpowder,M g Thickness of wick,T mm Height of total grooves,h t mm Spacing interval of grooves,s mm Depth of grooves,h g mm Width of grooves,w mm Porosity,e (%)Grooved ––– 1.360.80.650.85–Sintered 80–110 2.00 1.5––––55C40–6040–60 2.00 1.5 1.360.80.650.8554C60–8060–80 2.00 1.5 1.360.80.650.8554C80–11080–110 2.00 1.5 1.360.80.650.8554C110–140110–1402.001.51.360.80.650.8554Fig.3.Schematic of the risen meniscus testapparatus.Fig.4.Meniscus locating method of wick structure in IR images (sample:C110–140).192Y.Tang et al./Experimental Thermal and Fluid Science 34(2010)190–196in sintered porous layer.The interaction wetting repeats as the li-quid rises in wicks until a final equilibrium height reaches.As shown in Fig.5,the composite wick can be mainly divided into three parts for liquid flow,sintered porous zone,groove bot-tom zone and interaction wetting zone.Each flow zone provides a source of capillary force.Thus,we can obtain:D P cap ;com ¼D P cap ;sin þD P cap ;gro þD P cap ;int ð1Þwhere D P cap ,com is the capillary force of composite wicks,D P cap ,sin ,D P cap ,gro ,and D P cap ,int is defined to be the capillary force provided by sintered porous zone,groove bottom zone and interaction wet-ting zone,respectively.It should be noted that D P cap ,int is induced by the drag force of the fast liquid in groove,which acts on the slow liquid in sintered porous layer.4.Results and discussionsFor the rise of liquid during the risen meniscus test,capillary force of the wicks,D P cap ,plays the driving force.The total pressure loss,D P total ,plays the flow resistance,which consists of two parts,friction pressure loss and hydrostatic pressure loss,as follows:D P total ¼l e K h d hd tþq ghð2Þwhere l is the viscosity of working liquid,e is the porosity of wick structure,K is the permeability of wick structure,h is the wettedheight,d his the rising velocity of the wetted height,q is the liquid density,g is the gravitational acceleration.Because of the neglect of inertial effects,these are the only source of pressure change dur-ing the liquid rise.Thus,capillary force is equal to the total pressure loss:D P cap ¼l e h d hþq ghð3Þtherefore,the capillary force of different wicks can be compared byh and d hof liquid rise.For different wicks during the same test time,the sample with larger wetted height and rising velocity has larger capillary force.4.1.Effect of sintered layer on capillary forceAs can be seen in Fig.6,the liquid in all samples rose quickly at the early time of the rise process.As the wetted height increased,the rising velocity of the wetted height reduced -pared with the composite samples,the grooved sample had higher rising velocity in the first 15s.Without sintered powder in grooves,the friction resistance of liquid flow was small.Ethanol rose very fast in the grooves.The wetted height reached 30mm at 15s and the equilibrium height was obtained at 30s.After that the meniscus did not rise.While in the composite samples,the meniscus rose continuously for a longer time than grooved wick.The equilibrium height had not yet been obtained at the end of one minute.Fig.7shows the meniscus rising process of a compos-ite wick by IR thermal images.At the end of the test time,the wet-ted heights of all the four composite samples were larger than grooved wick.It can be concluded that composite wicks had larger capillary force than grooved one.This can be attributed to the anal-ysis in Wang’s report [3].Capillary force can be also obtained by the Laplace–Young equation as defined below:D P cap ¼2r r effð4Þwhere r is the surface tension of liquid and r eff is the effective cap-illary radius.As the sintered layer covered the grooves,a lot of small menisci formed in the sintered layer.The effective capillary radius decreased from the radius of big meniscus of the groove in grooved wick to the radius of small menisci in the sintered layer in compos-ite wick.Thus,from Eq.(4),the capillary force of composite wicks was larger than that of grooved one.Meanwhile,as analyzed in Sec-tion 3,from Eq.(1),sintered powder layer and interaction wetting provided two additional important portions of capillary force for composite wick.Thus,composite wicks have larger capillary force than grooved one.4.2.Effect of V-grooves on capillary forceA sintered sample,which has the same powder size with com-posite sample (80–110l m),was tested for comparison.The result was shown in Fig.8.During the first 10s,the wetted height andtheFig.5.Schematic of three liquid flow zones in a compositewick.Fig.6.Effects of sintered layer and powder size on wetted height of wicks.Y.Tang et al./Experimental Thermal and Fluid Science 34(2010)190–196193rising velocity were almost equal for both sintered and composite sample.It is due to that the capillary force difference between these two samples was small.Sintered porous layer in both wicks provided large capillary force for liquid rise.As the wetted height grew,the rising velocity reduced gradually.During this time,the capillary force difference between these two samples played an important role in liquid rise.Results show that the wetted height and the rising velocity of composite sample were larger than thatof sintered sample after 10s.It resulted from that the groove bot-tom zone and interaction wetting zone provided additional sources of capillary force for composite wick.Besides of the capillary force provided by sintered porous layer,the driving force of groove bot-tom zone and the drag force in interaction wetting zone increased the total capillary force of composite wicks.Thus,the capillary force of composite wick is a little larger than that of sintered wick.4.3.Effect of powder size on capillary force of composite wicks Experimental results in Fig.6shows that the wetted height of sample C80–110is the largest.Following was C110–140and C40–60.The smallest was C60–80.The rising velocity generally fol-lowed a similar order.Thus,the composite sample with the 80–110l m powder had the largest capillary force.It could be found that the capillary force of composite samples with large powder size (C80–110and C110–140)was larger than that with small powder size (C40–60and C60–80).This is mainly due to the porous matrix difference of composite wicks.As shown in Table 1,identi-cal porosity was obtained for these four samples.According to the definition of porosity [18],the volume sum of open pores and closed pores equaled for the samples,due to the same volume of the sintered layer in these four composite wicks.However,as shown in Fig.9,the open pore size and quantity varied,which strongly affected the liquid flow in wick structures.In porous ma-trix,only open pores could provide channels for liquid flow.Closed pores can not be saturated.Due to the existence of closed pores,li-quid would turn to flow by the adjacent open pores.Thus,theflow-Fig.7.IR images of the meniscus rising process of composite wick structure during one minute (sample:C110–140).parison between composite wick and sintered wick with the same powder size (80–110l m),wetted height versus time.194Y.Tang et al./Experimental Thermal and Fluid Science 34(2010)190–196ing channels became longer and the friction resistance of liquid flow increased.For the samples with small powder size (samples C40–60,C60–80),the voids among powder particles were apt to be filled or covered by other particles during sintering.A lot of closed pores existed.While for the samples with large powder size (samples C80–110,C110–140),there were larger voids among dif-ferent particles than that with small powder size.They could not be filled or covered by other particles due to their large size.More open pores formed.The open pores were bigger and they were eas-ier to interconnect with each other.Thus,there were more flowing channels for the liquid rise in sintered porous layer.The friction resistance of liquid flow was smaller.The liquid rose higher and faster.As a result,composite samples with large powder size have larger capillary force than that with small powder size.However,it was found in the experiments that capillary force does not increase linearly with powder size.It may be attributed to the difference of the interaction wetting effect.For the sample with large powder size (C80–110,C110–140),the interaction effect in sample C80–110with comparatively smaller powder size may be a little stronger in the liquid rise process,and the liquid in grooves may provide a little larger drag force acting on liquid in sintered porous layer than sample C110–140.Thus,the capillary force of sample C80–110is a little larger than sample C110–140.Similarly for the sample with small powder size,sample C40–60has slightly larger capillary force than sample C60–80.Other stud-ies,such as theoretical verification of powder size and other parameters’effects on interaction wetting,are still required in the future work.5.ConclusionsSintered–grooved composite wick has been presented for two-phase heat transfer devices.IR thermal imaging camera was usedin the risen meniscus test of capillary parison of the wetted height and rising velocity among composite,grooved and sintered wicks was conducted.Four composite wicks with different powder size ranged from 40–60l m to 110–140l m were tested.The conclusion can be summarized as follows:(1)For the infrared emissivity difference between liquid andsolid at a given temperature,the liquid meniscus in a wick can be accurately located by IR thermal images,including grooved,sintered and composite wicks.(2)Interaction wetting between groove and sintered powderhappens during the liquid rise,and composite wick can be mainly divided into three parts for liquid flow,sintered por-ous zone,groove bottom zone and interaction wetting zone.Each flow zone provides a source of capillary pos-ite wicks have larger capillary force than grooved and sin-tered ones.(3)Composite wicks with large powder size provide larger cap-illary force than that with small powder size.Friction resis-tance difference of liquid flow exists due to the variation in open pore size and quantity in different porous matrix of composite wicks.But the capillary force varies nonlinearly with the powder size.The optimal powder size of composite wick is suggested to be 80–110l m.Future work is needed for theoretically modeling of the liquid flow in composite wicks.Besides,theoretical analysis of powder size and other parameters’effect on interaction wetting and capil-lary force will be carried out.AcknowledgementsThis work is financially supported by the National Natural Sci-ence Foundation of China,Project No.U0834002andNo.Fig.9.SEM photograph of sintered porous matrix of composite samples showing variation in open pore size and quantity:(a–d)correspond to samples C40–60to C110–140,respectively.Y.Tang et al./Experimental Thermal and Fluid Science 34(2010)190–19619550705031,50975092,and Guangdong Natural Science Foundation, Project No.07118064,and No.8151064101000058.References[1]I.Sauciuc,M.Mochizuki,K.Mashiko,Y.Saito,T.Nguyen,The design and testingof the superfiber heat pipes for electronics cooling applications,in: Proceedings of16th IEEE Semiconductor Thermal Measurement and Management Symposium,San Jose,USA,2000,pp.27–32.[2]G.S.Hwang,M.Kaviany,W.G.Anderson,J.Zuo,Modulated wick heat pipe,Int.J.Heat Mass Transfer50(7–8)(2007)1420–1434.[3]J.L.Wang,I.Catton,Enhanced evaporation heat transfer in triangular groovescovered with a thinfine porous layer,Appl.Therm.Eng.21(17)(2001)1721–1737.[4]G.Franchi,X.Huang,Development of composite wicks for heat pipeperformance enhancement,Heat Transfer Eng.29(10)(2008)873–884. [5]D.R.Adkins,R.C.Dykhuizen,Procedures for measuring the properties of heatpipe wick materials,in:Proceedings of the28th Intersociety Energy Conversion Engineering Conference,Washington DC,1993,pp.911–917. [6]A.Das,A.K.Chatterjee,S.P.Basu,A method of measuring capillary rise in a heatpipe,Int.J.Heat Mass Transfer28(10)(1985)1959–1960.[7]A.Faghri,Heat Pipe Science and Technology,Taylor&Francis,Washington,DC,1995.[8]S.W.Chen,J.C.Hsieh,C.T.Chou,H.H.Lin,S.C.Shen,M.J.Tsai,Experimentalinvestigation and visualization on capillary and boiling limits of micro-grooves made by different processes,Sens.Actuat.A:Phys.139(1–2)(2007)78–87.[9]B.Holley, A.Faghri,Permeability and effective pore radius measurementsfor heat pipe and fuel cell applications,Appl.Therm.Eng.26(4)(2006) 448–462.[10]H.Kaplan,Practical applications of infrared thermal sensing and imageequipment,O’Shea Series Editor,Georgia Institute of Technology,USA, 1993.[11]R.Boukhanouf,A.Haddad,M.T.North,C.Buffone,Experimental investigationof aflat plate heat pipe performance using IR thermal imaging camera,Appl.Therm.Eng.26(17–18)(2006)2148–2156.[12]C.Buffone,K.Sefiane,Controlling evaporative thermocapillary convectionusing external heating:an experimental investigation,Exp.Therm.Fluid Sci.32(6)(2008)1287–1300.[13]M.Cerza,B.Boughey,The effects of air infiltration on a largeflat heat pipe athorizontal and vertical orientations,ASME J.Heat Transfer125(2)(2003)349–355.[14]D.T.Queheillalt,G.Carbajal,G.P.Peterson,H.N.G.Wadley,A multifunctionalheat pipe sandwich panel structure,Int.J.Heat Mass Transfer51(1-2)(2008) 312–326.[15]F.Zhang,J.Peng,J.Geng,Z.X.Wang,Z.B.Zhang,Thermal imaging study on thesurface wave of heated falling liquidfilms,Exp.Therm.Fluid Sci.33(3)(2009) 424–430.[16]D.F.Berry, E.Klar,Powder metal technologies and applications.In:ASMHandbook,vol.7.ASM International,Materials Park,USA,1998.p.309. [17]J.Lechman,N.Lu,Capillary force and water retention between two uneven-sized particles,J.Eng.Mech.ASCE134(5)(2008)74–384.[18]J.Banhart,Manufacture,characterisation and application of cellular metalsand metal foams,Prog.Mater.Sci.46(6)(2001)559–632.196Y.Tang et al./Experimental Thermal and Fluid Science34(2010)190–196。
激光衍射测定PMDI粒径(外文)
Respiratory Drug Delivery 2014 – Cooper et al.A QbD Method DevelopmentApproach for the Ex-actuatorParticle Size Distribution (PSD)Determination of pMDIsby Laser DiffractionAndy Cooper,1 Chris Blatchford,1and Stephen Stein213M Drug Delivery Systems Ltd, Loughborough,Leicestershire,UK23M Drug Delivery Systems, St. Paul, MN, USAKEYWORDS: pressurized metered dose inhaler (pMDI),quality by design (QbD), laser diffraction,method development, particle sizeIntRODuCtIOnThe aerodynamic particle size distribution (APSD) is a critical quality attribute (CQA) of orally inhaled and nasal drug products (OINDPs). Cascade impactor methodology [1] is typically used to determine this during development and registered product testing. However, this is a time consuming analysis which makes it impractical as a rapid screening tool for early phase develop-ment and investigations. Laser diffraction (LD) analysis of the ex-actuator plume has previously been proposed as an alternative [2] for geometric particle size distribution (PSD) determination. This is considered appropriate as the measured ex-actuator PSD would be influenced by the active pharmaceutical ingredient (API) within the formulation in three ways. One, the dry API par-ticles would be present in the actuation plume after atomization. Two, droplet formation during atomization may be influenced by the particle size of the API [3]. Lastly, the API concentration will influence the number of particles within each droplet and hence the measured PSD [2, 4]. This publication will focus on the specific considerations for method development/validation of this methodology [5-11], using a quality by design (QbD) approach [11, 12]. Data from these methods have the potential to correlate with APSD data [3, 14], as many variables are common to both techniques [15], despite the difference in the principle of measurement (geometric versus aerodynamic).321322QbD Method Development for the Ex-actuator PSD of pMDIs by Laser Diffraction – Cooper et al.MEthODOLOgYMeasurements were made with a Sympatec™ instrument, consisting of a Helos (Helium-Neon Laser Optical System)/BF™ laser diffraction sensor with the Inhaler™ dry dispersion accessory (See instrument settings in Table 1). The pMDI unit is manually actuated into the Inhaler accessory, allowing the actuation plume to pass through the laser (He/Ne @ 633nm) where light is scattered and then focused by a chosen lens onto a detector array. The detector consists of 31 concentric ring elements. The innermost element is referred to as R1 and the outermost element as R31. Scattered light registered on elements R1-R6 is discounted due to beam steer [16]. The signals from all the other detector elements are combined and a PSD is inferred from the scattering pattern based on the chosen optical model, after subtraction of background levels.table 1.Sympatec™ instrument settings.RESuL tS AnD DISCuSSIOnThere are multiple variables which can influence laser diffraction data, as shown in Figure 1. Aside from the product related factors, risk assessments of the other variables are required. While some parameters have no impact if they are suitably controlled (e.g., cleaning), other more critical parameters require experimenta-tion to determine their effects (e.g., trigger conditions). The Inhaler accessory was chosen since the droplets are entrained in an air flow which dilutes the sample and reduces potential artefacts such as velocity bias and geometric effects [15].Figure 1. Fishbone diagram of sources of variability for laser diffraction data.Respiratory Drug Delivery 2014 – Cooper et al. 323Copyright © 2014 VCUThe length of cylinder used for the Inhaler accessory must be optimized. This dictates thedistance between the point of actuation and the laser beam. The optical concentration (OC) must be sufficient for sensitivity purposes but not too high to cause multiple scattering. Data are shown in Figure 2A and 2B. The similarity in OC for the active and placebo shows that the majority of light did not reach the central detector due to the propellant shifting the laser beam (beam steer). The median particle size for the placebo is decreased with increased distance. This is likely due to the increased evaporation of co-solvent, rather than an indication of multiple scattering. This signal observed for the placebo led to the refractive index of the co-solvent being chosen for the method. Although the OC decreased with increased firing distance due to reduced beam steer, the active PSD remains reasonably consistent [<10% shift in d(v, 0.5)] and is considered to be real. A slight increase in active median particle size is consistent with reduced interference from the co-solvent and therefore the long cylinder is considered to be the most accurate.Figure 2. A: Optical concentration. B: d(v, 0.5) data for high strength active and placebo for various cylinder lengths.The selection of sampling criteria (trigger conditions) is critical [4] and they can alsointeract with the method of shaking and firing, particularly for a suspension pMDI. Design experi-ments are crucial for method optimization and for proving method robustness. Parameters, which are selected via a suitable risk assessment, are evaluated over an appropriate design space, as shownin Figure 3.324QbD Method Development for the Ex-actuator PSD of pMDIs by Laser Diffraction – Cooper et al.Figure 3. PSD data for sample preparation DoE – low strength active.Validation data for this method are shown in Table 2. Validation was performed on both high and low strength actives. RSD values are higher than those typically observed for a laser diffraction method for an API [17], however this is likely due to the inherent product variability (See Figure 1) – measurements are made of volatile droplets with a wide range of velocities contributing to differing amounts of evaporation per droplet. Increased variability is therefore expected. However, one important factor to limit variability is humidity level [4].table 2.Method validation data.This validated method has been used to try and understand trends in APSD data. The data shown in Table 3 shows that the PSD is influenced by the temperature of the product, which has similar trends to those observed for ACI data in the literature [18]. The vapor pressure of the formulation increases with temperature, which results in atomization of smaller droplets, which will contain fewer drug particles, hence a decrease in PSD. Faster propellant/co-solvent evaporation, due to the increased temperature, will also result in a decrease in PSD.table 3.PSD data for a range of temperatures (n = 6 at each condition) – high strength active.Respiratory Drug Delivery 2014 – Cooper et al. 325COnCLuSIOnSA QbD approach to development of laser diffraction (LD) methodology to determine the PSD ex-actuator from a pressurized metered dose inhaler (pMDI) is presented. Robust and repeatable data were obtained, however this is inherently more variable than methodology for the PSD deter-mination of the API. The LD methodology can be used to understand trends in cascade impaction data – the industry standard, but more time consuming, methodology for APSD determination.REFEREnCES1. USP Chapter (601): Aerosols, nasal sprays, metered-dose inhalers, and dry powder inhalers.2. Gonda, I: Development of a systematic theory of suspension inhalation aerosols. A frameworkto study the effects of aggregation on the aerodynamic behaviour of drug particles, International Journal of Pharmaceutics 1985, 27: 99-116.3. Pu, Y, Kline, LC, Berry, J: The application of “in-flight” laser diffraction to the particle sizecharacterization of a model suspension metered dose inhaler, Drug Development and Industrial Pharmacy 2011, 37 (5): 552-58.4. Cooper, A, Bell, T: Monitoring of droplet size changes in a suspension pMDI by laserdiffraction on a Sympatec™Instrument. In Drug Delivery to the Lungs 19, 2008.5. Mitchell, JP, Nagel, MW, Nichols, S, Nerbrink, O: Laser diffractometry as a technique for therapid assessment of aerosol particle size from inhalers, Journal of Aerosol Medicine 2006, 19(4): 409-33.6. ICH Harmonised Tripartite Guideline Q2(R1) (1994): Validation of analytical procedures,Text and Methodology.7. ISO Standard 13320-1 (2009): Particle size analysis, laser diffraction methods.8. Ph. Eur. Chapter 2.9.31: Particle size analysis by laser light diffraction.9. USP Chapter (429): Light diffraction measurement of particle size.10. Ward-Smith, RS, Gummery, N, Rawle, AF: Validation of wet and dry laser diffraction particlecharacterisation methods. Malvern Instruments Ltd. /malvern/kbase.nsf/allbyno/KB000167/$file/Laser%20Diffraction%20Method%20Validation.pdf.11. Rawle, A, Kippax, P: Setting new standards for laser diffraction particle size analysis. MalvernInstruments Ltd. /malvern/kbase.nsf/allbyno/KB002403/$file MRK 1399-01.pdf.Copyright © 2014 VCU326QbD Method Development for the Ex-actuator PSD of pMDIs by Laser Diffraction – Cooper et al.12. Schweitzer, M, Pohl, M, Hanna-Brown, M, Nethercote, P, Borman, P, Hansen, G, Smith,K, Wegener, G: I mplications and opportunities of applying QbD principles to analytical measurements, Pharmaceutical Technology 2010, 34 (2): 52-59.13. Borman, P, Chatfield, M, Jackson, P, Laures, A, Okafo, G: Reduced-method robustness testingof analytical methods driven by a risk-based approach, Pharmaceutical Technology 2010, 34(4): 72-86.14. Jones, SA, Martin, GP, Brown, M.B: High-pressure aerosol suspensions – A novel laserdiffraction particle sizing system for hydrofluoroalkane pressurised metered dose inhalers, International Journal of Pharmaceutics 2005, 302: 154-65.15. Blatchford, C: From powder to patient - optimisation of particle sizing techniques, In DrugDelivery to the Lungs 24, 2013.16. Ranucci, J: Dynamic plume – particle size analysis using laser diffraction, PharmaceuticalTechnology 1992, 16: 108-14.17. Cooper, A, Blatchford, C, Kelly, M: Laser diffraction methodology for particle size distribution(PSD) determination during pMDI product development – A QbD approach. In Respiratory Drug Delivery Europe 2013. Volume 2. Edited by Dalby, RN, Byron, PR, Peart, J, Suman, JD, Young, PM, Traini, D. DHI Publishing; River Grove, IL: 2013: 197-202.18. Stein, S, Cocks, P: Size distribution measurements from metered dose inhalers at lowtemperatures. I n Respiratory Drug Delivery Europe 2013. Volume 2. Edited by Dalby, RN, Byron, PR, Peart, J, Suman, JD, Young, PM, Traini, D. DHI Publishing; River Grove, IL: 2013: 203-08.。
liquid–liquid phase separation
Biophysical Chemistry 109(2004)105–1120301-4622/04/$-see front matter ᮊ2003Elsevier B.V .All rights reserved.doi:10.1016/j.bpc.2003.10.021Cloud-point temperature and liquid–liquid phase separation ofsupersaturated lysozyme solutionJie Lu *,Keith Carpenter ,Rui-Jiang Li ,Xiu-Juan Wang ,Chi-Bun Ching a ,a a b bInstitute of Chemical and Engineering Sciences,Ayer Rajah Crescent 28,࠻02-08,Singapore 139959,Singapore aChemical and Process Engineering Center,National University of Singapore,Singapore 117576,SingaporebReceived 31July 2003;received in revised form 8October 2003;accepted 16October 2003AbstractThe detailed understanding of the structure of biological macromolecules reveals their functions,and is thus important in the design of new medicines and for engineering molecules with improved properties for industrial applications.Although techniques used for protein crystallization have been progressing greatly,protein crystallization may still be considered an art rather than a science,and successful crystallization remains largely empirical and operator-dependent.In this work,a microcalorimetric technique has been utilized to investigate liquid–liquid phase separation through measuring cloud-point temperature T for supersaturated lysozyme solution.The effects of cloud ionic strength and glycerol on the cloud-point temperature are studied in detail.Over the entire range of salt concentrations studied,the cloud-point temperature increases monotonically with the concentration of sodium chloride.When glycerol is added as additive,the solubility of lysozyme is increased,whereas the cloud-point temperature is decreased.ᮊ2003Elsevier B.V .All rights reserved.Keywords:Biocrystallization;Microcalorimetry;Cloud-point temperature;Liquid–liquid phase separation1.IntroductionKnowledge of detailed protein structure is essen-tial for protein engineering and the design of pharmaceuticals.Production of high-quality pro-tein crystals is required for molecular structure determination by X-ray crystallography.Although considerable effort has been made in recent years,obtaining such crystals is still difficult in general,and predicting the solution conditions where pro-*Corresponding author.Tel.:q 65-6874-4218;fax:q 65-6873-4805.E-mail address:lujie@.sg (J.Lu ).teins successfully crystallize remains a significant obstacle in the advancement of structural molecu-lar biology w 1x .The parameters affecting protein crystallization are typically reagent concentration,pH,tempera-ture,additive,etc.A phase diagram can provide the method for quantifying the influence of solu-tion parameters on the production of crystals w 2,3x .To characterize protein crystallization,it is neces-sary to first obtain detailed information on protein solution phase behavior and phase diagram.Recently physics shows that there is a direct relationship between colloidal interaction energy106J.Lu et al./Biophysical Chemistry109(2004)105–112and phase diagram.Gast and Lekkerkerker w4,5x have indicated that the range of attraction between colloid particles has a significant effect on the qualitative features of phase diagram.A similar relationship should hold for biomacromolecules, i.e.the corresponding interaction potentials govern the macromolecular distribution in solution,the shape of the phase diagram and the crystallization process w6x.Many macromolecular crystallizations appear to be driven by the strength of the attractive interactions,and occur in,or close to,attractive regimes w7,8x.Recent intensive investigation has revealed that protein or colloidal solution possesses a peculiar phase diagram,i.e.liquid–liquid phase separation and sol–gel transition exists in general in addition to crystallization w9,10x.The potential responsible for the liquid–liquid phase separation is a rather short range,possibly van der Waals,attractive potential w11,12x.The measurement of cloud-point temperature T can provide useful informationcloudon the net attractive interaction between protein molecules,namely,the higher the cloud-point tem-perature,the greater the net attractive interaction. Herein Taratuta et al.w13x studied the effects of salts and pH on the cloud-point temperature of lysozyme.Broide et al.w14x subsequently meas-ured the cloud-point temperature and crystalliza-tion temperature for lysozyme as a function of salt type and concentration.From these works the cloud-point temperature was found to be typically 15–458C below the crystallization temperature. Furthermore,Muschol and Rosenberger w15x deter-mined the metastable coexistence curves for lyso-zyme through cloud-point measurements,and suggested a systematic approach to promote pro-tein crystallization.In general,an effective way to determine the strength of protein interactions is to study temperature-induced phase transitions that occur in concentrated protein solutions.Liquid–liquid phase separation can be divided into two stages w11x:(1)the local separation stage at which the separation proceeds in small regions and local equilibrium is achieved rapidly;and(2) the coarsening stage at which condensation of these small domains proceeds slowly to reduce the loss of interface free energy w16x.The coexisting liquid phases both remain supersaturated but differ widely in protein concentration.The effect of a metastable liquid–liquid phase separation on crystallization remains ambiguous w17x.Molecular dynamics simulations and analyt-ical theory predict that the phase separation will affect the kinetics and the mechanisms of protein crystal nucleation w18x.tenWolde and Frenkel w19x have demonstrated that the free energy barrier for crystal nucleation is remarkably reduced at the critical point of liquid–liquid phase separation, thus in general,after liquid–liquid phase separa-tion,crystallization occurs much more rapidly than in the initial solution,which is typically too rapid for the growth of single crystal with low defect densities w15x.The determination of the location of liquid–liquid phase separation curve is thus crucial for efficiently identifying the optimum solution conditions for growing protein crystals. Microcalorimetry has the potential to be a useful tool for determining:(1)the metastable-labile zone boundary;(2)the temperature-dependence of pro-tein solubility in a given solvent;and(3)the crystal-growth rates as a function of supersatura-tion w20x.Microcalorimeters can detect a power signal as low as a few microwatts whereas standard calorimeters detect signals in the milliwatt range. Because of this greater sensitivity,samples with small heat effects can be analyzed.In addition, microcalorimetry has the advantage of being fast, non-destructive to the protein and requiring a relatively small amount of material.The present work is concerned with the analysis of the transient heat signal from microcalorimeter to yield liquid–liquid phase separation information for lysozyme solutions at pH4.8.To further examine the role of salt and additive on interprotein interactions, cloud-point temperature T has been determinedcloudexperimentally as a function of the concentrations of salt,protein and glycerol.2.Materials and methods2.1.MaterialsSix times crystallized lysozyme was purchased from Seikagaku Kogyo,and used without further107J.Lu et al./Biophysical Chemistry 109(2004)105–112purification.All other chemicals used were of reagent grade,from Sigma Chemical Co.2.2.Preparation of solutionsSodium acetate buffer (0.1M )at pH 4.8was prepared with ultrafiltered,deionized water.Sodi-um azide,at a concentration of 0.05%(w y v ),was added to the buffer solution as an antimicrobial agent.Protein stock solution was prepared by dissolving protein powder into buffer.To remove undissolved particles,the solution was centrifuged in a Sigma centrifuge at 12000rev.y min for 5–10min,then filtered through 0.22-m m filters (Mil-lex-VV )into a clean sample vial and stored at 48C for further experiments.The concentration of protein solution was determined by measuring the absorbance at 280nm of UV spectroscopy (Shi-madzu UV-2550),with an extinction coefficient of 2.64ml y (mg cm )w 21x .Precipitant stock solution was prepared by dissolving the required amount of sodium chloride together with additive glycerol into buffer.The pH of solutions was measured by a digital pH meter (Mettler Toledo 320)and adjusted by the addition of small volumes of NaOH or HAc solution.2.3.Measurement of solubilitySolubility of lysozyme at various temperatures and precipitant y additive concentrations was meas-ured at pH 4.8in 0.1M acetate buffer.Solid–liquid equilibrium was approached through both crystallization and dissolution.Dissolving lasted 3days,while the period of crystallization was over 2weeks.The supernatant in equilibrium with a macroscopically observable solid was then filtered through 0.1-m m filters (Millex-VV ).The concen-tration of diluted supernatant was determined spec-troscopically and verified by refractive meter(Kruss)until refractive index remained unchanged ¨at equilibrium state.Solubility of each sample was measured in duplicate.2.4.Differential scanning microcalorimetry Calorimetric experiments were performed with a micro-differential scanning calorimeter with anultra sensitivity,micro-DSC III,from Setaram SA,France.The micro-DSC recorded heat flow in microwatts vs.temperature,thus can detect the heat associated with phase transition during a temperature scan.The sample made up of equal volumes of protein solution and precipitant solu-tion was filtered through 0.1-m m filters to remove dust particles further.To remove the dissolved air,the sample was placed under vacuum for 3min while stirring at 500rev.y min by a magnetic stirrer.The degassed sample was placed into the sample cell of 1.0ml,and a same concentration NaCl solution was placed into the reference cell.The solutions in the micro-DSC were then cooled at the rate of 0.28C y min.After every run,the cells were cleaned by sonicating for 10–15min in several solutions in the following order:deionized water,methanol,ethanol,acetone,1M KOH and finally copious amounts of deionized water.This protocol ensured that lysozyme was completely removed from the cells.The cells were then placed in a drying oven for several hours.The rubber gaskets were cleaned in a similar manner except acetone and 1M KOH were omitted and they were allowed to dry at low temperature.3.Results and discussionA typical micro-DSC scanning experiment is shown in Fig.1.The onset of the clouding phe-nomenon is very dramatic and easily detected.The sharp increase in the heat flow is indicative of a liquid–liquid phase separation process producing a latent heat.This is much consistent with many recent investigations of the liquid–liquid phase separation of lysozyme from solution w 22,23x .In fact,such a liquid–liquid phase separation is a phase transition with an associated latent heat of demixing.In this work,the cloud-point tempera-tures at a variety of lysozyme,NaCl and glycerol concentrations are determined by the micro-DSC at the scan rate of 128C y h.3.1.Effect of protein concentrationIn semilogarithmic Fig.2we plot the solid–liquid and liquid–liquid phase boundaries for lyso-108J.Lu et al./Biophysical Chemistry 109(2004)105–112Fig.1.Heat flow of a typical micro-DSC scan of lysozyme solution,50mg y ml,0.1M acetate buffer,pH 4.8,3%NaCl.The scan rate 128C y h is chosen referenced to the experimental results of Darcy and Wiencek w 23x .Note the large deflection in the curve at approximately 4.38C indicating a latent heat resulting from demixing (i.e.liquid–liquid phase separation )process.Fig.2.Cloud-point temperature and solubility determination for lysozyme in 0.1M acetate buffer,pH 4.8:solubility (5%NaCl )(s );T (5%NaCl,this work )(d );T (5%cloud cloud NaCl,the work of Darcy and Wiencek w 23x )(*);solubility (3%NaCl )(h );T (3%NaCl )(j ).cloud Fig.3.Cloud-point temperature determination for lysozyme as a function of the concentration of sodium chloride,50mg y ml,0.1M acetate buffer,pH 4.8.zyme in 0.1M acetate buffer,pH 4.8,for a range of protein concentrations.It is worth noting that,at 5%NaCl,our experimental data of T from cloud micro-DSC are quite consistent with those from laser light scattering and DSC by Darcy and Wiencek w 23x ,with difference averaging at approx-imately 0.88C.This figure demonstrates that liquid–liquid phase boundary is far below solid–liquid phase boundary,which implies that the liquid–liquid phase separation normally takes place in a highly metastable solution.In addition,cloud-point temperature T increases with the cloud concentration of protein.3.2.Effect of salt concentrationFig.3shows how cloud-point temperature changes as the concentration of NaCl is varied from 2.5to 7%(w y v ).The buffer is 0.1M acetate (pH 4.8);the protein concentration is fixed at 50mg y ml.Over the entire range of salt concentrations studied,the cloud-point temperature strongly depends on the ionic strength and increases monotonically with the concentration of NaCl.Crystallization is driven by the difference in chemical potential of the solute in solution and in the crystal.The driving force can be simplified as w 24xf sy Dm s kT ln C y C (1)Ž.eq109J.Lu et al./Biophysical Chemistry 109(2004)105–112Fig.4.The driving force required by liquid–liquid phase sep-aration as a function of the concentration of sodium chloride,50mg y ml lysozyme solution,0.1M acetate buffer,pH 4.8.In the same way,we plot the driving force,f ,required by liquid–liquid phase separation as a function of the concentration of sodium chloride in Fig.4.At the moderate concentration of sodium chloride,the driving force required by liquid–liquid phase separation is higher than that at low or high salt concentration.As shown in Fig.3,with NaCl concentration increasing,the cloud-point temperature increases,which is in accord with the results of Broide et al.w 14x and Grigsby et al.w 25x .It is known that protein interaction is the sum of different potentials like electrostatic,van der Waals,hydrophobic,hydration,etc.The liquid–liquid phase separation is driven by a net attraction between protein molecules,and the stronger the attraction,the higher the cloud-point temperature.Ionic strength is found to have an effect on the intermolecular forces:attractions increase with ionic strength,solubility decreases with ionic strength,resulting in the cloud-point temperature increases with ionic strength.It is worth noting that,the effect of ionic strength on cloud-point temperature depends strongly on the specific nature of the ions w 13x .Kosmotropic ions bind adjacent water molecules more strongly than water binds itself.When akosmotropic ion is introduced into water,the entro-py of the system decreases due to increased water structuring around the ion.In contrast,chaotropes bind adjacent water molecules less strongly than water binds itself.When a chaotrope is introduced into water,the entropy of the system increases because the water structuring around the ion is less than that in salt-free water.This classification is related to the size and charge of the ion.At high salt concentration ()0.3M ),the specific nature of the ions is much more important w 25x .The charges on a protein are due to discrete positively and negatively charged surface groups.In lysozyme,the average distance between thesecharges is approximately 10Aw 26x .As to the salt ˚NaCl used as precipitant,Na is weakly kosmo-q tropic and Cl is weakly chaotropic w 27x .At low y NaCl concentrations,as the concentration of NaCl increases,the repulsive electrostatic charge–charge interactions between protein molecules decrease because of screening,resulting in the increase of cloud-point temperature.While at high NaCl con-centrations,protein molecules experience an attrac-tion,in which differences can be attributed to repulsive hydration forces w 14,25x .That is,as the ionic strength increases,repulsive electrostatic or hydration forces decrease,protein molecules appear more and more attractive,leading to higher cloud-point temperature.At various salt concentra-tions,the predominant potentials reflecting the driving force for liquid–liquid phase separation are different.Fig.4shows that the driving force,f ,is parabolic with ionic strength,while Grigsby et al.w 25x have reported that f y kT is linear with ionic strength for monovalent salts.The possible reasons for that difference include,their model is based on a fixed protein concentration of 87mg y ml,which is higher than that used in our study,yet f y kT is probably dependent on protein concentration,besides the solutions at high protein and salt concentrations are far from ideal solutions.3.3.Effect of glycerolFig.5compares cloud-point temperature data for 50mg y ml lysozyme solutions in absence of glycerol and in presence of 5%glycerol,respec-110J.Lu et al./Biophysical Chemistry109(2004)105–112parison of cloud-point temperatures for lysozyme at different glycerol concentrations as a function of the con-centration of sodium chloride,50mg y ml,0.1M acetate buffer, pH4.8:0%glycerol(s);5%glycerol(j).Fig.6.Cloud-point temperatures for lysozyme at different glycerol concentrations,50mg y ml lysozyme,5%NaCl,0.1M acetate buffer,pH4.8.Fig.7.Cloud-point temperature and solubility determination for lysozyme at different concentrations of glycerol in0.1M acetate buffer,5%NaCl,pH4.8:solubility(0%glycerol)(s); T(0%glycerol)(d);solubility(5%glycerol)(h);cloudT(5%glycerol)(j).cloudtively.Fig.6shows the cloud-point temperature as a function of the concentration of glycerol.The cloud-point temperature is decreased as the addi-tion of glycerol.In semilogarithmic Fig.7we plot the solid–liquid and liquid–liquid phase boundaries at dif-ferent glycerol concentrations for lysozyme in0.1 M acetate buffer,5%NaCl,pH4.8,for a range of protein concentration.This figure demonstrates that liquid–liquid and solid–liquid phase bounda-ries in the presence of glycerol are below those in absence of glycerol,and the region for growing crystals is narrowed when glycerol is added. Glycerol has the property of stabilizing protein structure.As a result,if crystallization occurs over a long period of time,glycerol is a useful candidate to be part of the crystallization solvent and is often included for this purpose w28x.In addition,glycerol is found to have an effect on the intermolecular forces:repulsions increase with glycerol concentra-tion w29x.Our experiment results of solubility and cloud-point temperature can also confirm the finding.The increased repulsions induced by glycerol can be explained by a number of possible mecha-nisms,all of which require small changes in the protein or the solvent in its immediate vicinity.The addition of glycerol decreases the volume of protein core w30x,increases hydration and the size of hydration layer at the particle surface w31,32x. In this work,we confirm that glycerol shifts the solid–liquid and liquid–liquid phase boundaries. The effect of glycerol on the phase diagram strong-111 J.Lu et al./Biophysical Chemistry109(2004)105–112ly depends on its concentration and this canprovide opportunities for further tuning of nuclea-tion rates.4.ConclusionsGrowing evidence suggests protein crystalliza-tion can be understood in terms of an order ydisorder phase transition between weakly attractiveparticles.Control of these attractions is thus keyto growing crystals.The study of phase transitionsin concentrated protein solutions provides one witha simple means of assessing the effect of solutionconditions on the strength of protein interactions.The cloud-point temperature and solubility datapresented in this paper demonstrate that salt andglycerol have remarkable effects on phase transi-tions.The solid–liquid and liquid–liquid bounda-ries can be shifted to higher or lower temperaturesby varying ionic strength or adding additives.Ourinvestigation provides further information upon therole of glycerol used in protein crystallization.Glycerol can increase the solubility,and decreasethe cloud-point temperature,which is of benefit totuning nucleation and crystal growth.In continuingstudies,we will explore the effects of other kindsof additives like nonionic polymers on phasetransitions and nucleation rates.Much more theo-retical work will be done to fully interpret ourexperimental results.AcknowledgmentsThis work is supported by the grant from theNational Natural Science Foundation of China(No.20106010).The authors also thank Professor J.M.Wiencek(The University of Iowa)for kinddiscussion with us about the thermal phenomenaof liquid–liquid phase separation.Referencesw1x A.McPherson,Current approaches to macromolecular crystallization,Eur.J.Biochem.189(1990)1–23.w2x A.M.Kulkarni, C.F.Zukoski,Nanoparticle crystal nucleation:influence of solution conditions,Langmuir18(2002)3090–3099.w3x E.E.G.Saridakis,P.D.S.Stewart,L.F.Lloyd,et al., Phase diagram and dilution experiments in the crystal-lization of carboxypeptidase G2,Acta Cryst.D50(1994)293–297.w4x A.P.Gast, C.K.Hall,W.B.Russel,Polymer-induced phase separations in non-aqueous colloidal suspensions,J.Colloid Interf.Sci.96(1983)251–267.w5x H.N.W.Lekkerkerker,W.C.K.Poon,P.N.Pusey,et al., Phase-behavior of colloid plus polymer mixtures,Euro-phys.Lett.20(1992)559–564.w6x A.Tardieu,S.Finet,F.Bonnete,Structure of the´macromolecular solutions that generate crystals,J.Cryst.Growth232(2001)1–9.w7x D.Rosenbaum,C.F.Zukoski,Protein interactions and crystallization,J.Cryst.Growth169(1996)752–758.w8x A.George,W.W.Wilson,Predicting protein crystalli-zation from a dilute solution property,Acta Cryst.D50(1994)361–365.w9x D.Rosenbaum,P.C.Zamora, C.F.Zukoski,Phase-behavior of small attractive colloidal particles,Phys.Rev.Lett.76(1996)150–153.w10x V.J.Anderson,H.N.W.Lekkerkerker,Insights into phase transition kinetics from colloid science,Nature416(2002)811–815.w11x S.Tanaka,K.Ito,R.Hayakawa,Size and number density of precrystalline aggregates in lysozyme crys-tallization process,J.Chem.Phys.111(1999)10330–10337.w12x D.W.Liu,A.Lomakin,G.M.Thurston,et al.,Phase-separation in multicomponent aqueous-protein solutions,J.Phys.Chem.99(1995)454–461.w13x V.G.Taratuta,A.Holschbach,G.M.Thurston,et al., Liquid–liquid phase separation of aqueous lysozymesolutions:effects of pH and salt identity,J.Phys.Chem.94(1990)2140–2144.w14x M.L.Broide,T.M.Tominc,M.D.Saxowsky,Using phase transitions to investigate the effect of salts onprotein interactions,Phys.Rev.E53(1996)6325–6335. w15x M.Muschol,F.Rosenberger,Liquid–liquid phase sep-aration in supersaturated lysozyme solutions and asso-ciated precipitate formation y crystallization,J.Chem.Phys.107(1997)1953–1962.w16x C.Domb,J.H.Lebowitz,Phase Separation and Critical Phenomena,Academic,London,1983.w17x D.F.Rosenbaum,A.Kulkarni,S.Ramakrishnan,C.F.Zukoski,Protein interactions and phase behavior:sen-sitivity to the form of the pair potential,J.Chem.Phys.111(1999)9882–9890.w18x O.Galkin,P.G.Vekilov,Nucleation of protein crystals: critical nuclei,phase behavior and control pathways,J.Cryst.Growth232(2001)63–76.w19x P.R.tenWolde, D.Frenkel,Enhancement of protein crystal nucleation by critical density fluctuations,Sci-ence277(1997)1975–1978.w20x P.A.Darcy,J.M.Wiencek,Estimating lysozyme crystal-lization growth rates and solubility from isothermalmicrocalorimetry,Acta Cryst.D54(1998)1387–1394.112J.Lu et al./Biophysical Chemistry109(2004)105–112w21x A.J.Sophianopoulos,C.K.Rhodes,D.N.Holcomb,K.E.vanHolde,Physical studies of lysozyme.I.Characteri-zation,J.Biol.Chem.237(1962)1107–1112.w22x Y.Georgalis,P.Umbach, A.Zielenkiewicz,et al., Microcalorimetric and small-angle light scattering stud-ies on nucleating lysozyme solutions,J.Am.Chem.Soc.119(1997)11959–11965.w23x P.A.Darcy,J.M.Wiencek,Identifying nucleation tem-peratures for lysozyme via differential scanning calorim-etry,J.Cryst.Growth196(1999)243–249.w24x M.L.Grant,Effects of thermodynamics nonideality in protein crystal growth,J.Cryst.Growth209(2000)130–137.w25x J.J.Grigsby,H.W.Blanch,J.M.Prausnitz,Cloud-point temperatures for lysozyme in electrolyte solutions:effectof salt type,salt concentration and pH,Biophys.Chem.91(2001)231–243.w26x D.Voet,J.Voet,Biochemistry,Wiley,New Y ork,1990. w27x K.D.Collins,Charge density-dependent strength of hydration and biological structure,Biophys.J.72(1997)65–76.w28x R.Sousa,Use of glycerol and other protein structure stabilizing agents in protein crystallization,Acta Cryst.D51(1995)271–277.w29x M.Farnum, C.F.Zukoski,Effect of glycerol on the interactions and solubility of bovine pancreatic trypsininhibitor,Biophys.J.76(1999)2716–2726.w30x A.Priev,A.Almagor,S.Yedgar,B.Gavish,Glycerol decreases the volume and compressibility of proteininterior,Biochemistry35(1996)2061–2066.w31x S.N.Timasheff,T.Arakawa,Mechanism of protein precipitation and stabilization by co-solvents,J.Cryst.Growth90(1988)39–46.w32x C.S.Miner,N.N.Dalton,Glycerol,Reinhold Publishing, New Y ork,1953.。
2007年诺贝尔物理学奖
2007年诺贝尔物理学奖2007年物理学奖，由两位物理学家分享，他们是法国的艾尔伯·费尔（Albert Fert）和德国的皮特·克鲁伯格（Peter Grünberg）。
他们于1988年，各自独立地发现了巨磁电阻效应，极大地提高了电脑硬盘的数据存储量。
艾尔伯·费尔（Albert Fert，1938—），出生于法国的卡尔卡松。
1962年，费尔在巴黎高等师范学院获得数学和物理硕士学位。
1970年，费尔从巴黎第十一大学获得物理学博士学位，并留校任教。
费尔从1970年到1995年一直在巴黎第十一大学固体物理实验室工作，后任研究小组组长。
1995年至今则担任国家科学研究中心-Thales集团联合物理小组科学主管。
1988年，费尔发现巨磁电阻效应，同时他对自旋电子学作出过许多贡献。
皮特·克鲁伯格（Peter Grünberg，1939—2018），出生于德国。
从1959年到1963年，克鲁伯格在法兰克福约翰-沃尔夫冈-歌德大学学习物理，1962年获得中级文凭，1969年在达姆施塔特技术大学获得博士学位。
1988年，克鲁伯格在尤利西研究中心研究并发现巨磁电阻效应。
1992年被任命为科隆大学兼任教授。
2004年在研究中心工作32年后退休，但仍继续工作，直到2018年逝世。
巨磁电阻效应是指当铁磁材料（Ferromagnetic）和非磁性金属（Non-Magnetic Metal）层交替组合成的材料，在既使微弱的磁场作用下铁磁层的电阻突然巨幅下降的现象。
特别值得注意的是，如果相邻材料中铁磁层的磁化方向平行的1时候，电阻会变得很低；而当铁磁层的磁化方向相反的时候电阻则会变得很大。
电阻值的这种变化是由于不同自旋的电子在单层磁化材料中的散射性质不同而造成的。
早在1988年，费尔和克鲁伯格就各自独立发现了这一特殊现象：非常弱小的磁场变化就能导致磁性材料发生非常显著的电阻变化。
Discrete Applied Mathematics
Discrete Applied Mathematics157(2009)2217–2220Contents lists available at ScienceDirectDiscrete Applied Mathematicsjournal homepage:/locate/damPreface$This special issue on Networks in Computational Biology is based on a workshop at Middle East Technical University in Ankara,Turkey,September10–12,2006(.tr/Networks_in_Computational_Biology/). Computational biology is one of the many currently emerging areas of applied mathematics and science.During the last century,cooperation between biology and chemistry,physics,mathematics,and other sciences increased dramatically,thus providing a solid foundation for,and initiating an enormous momentum in,many areas of the life sciences.This special issue focuses on networks,a topic that is equally important in biology and mathematics,and presents snapshots of current theoretical and methodological work in network analysis.Both discrete and continuous optimization,dynamical systems, graph theory,pertinent inverse problems,and data mining procedures are addressed.The principal goal of this special issue is to contribute to the mathematical foundation of computational biology by stressing its particular aspects relating to network theory.This special issue consists of25articles,written by65authors and rigorously reviewed by70referees.The guest editors express their cordial thanks to all of them,as well as to the Editors-in-Chief of Discrete Applied Mathematics,Prof.Dr.Endre Boros and his predecessor,Prof.Dr.Peter L.Hammer,who was one of the initiators of this special issue but left us in2006, and to Mrs.Katie D’Agosta who was at our side in each phase of preparation of this DAM special issue.The articles are ordered according to their contents.Let us briefly summarize them:In the paper of Jacek Błażewicz,Dorota Formanowicz,Piotr Formanowicz,Andrea Sackmann,and MichałSajkowski, entitled Modeling the process of human body iron homeostasis using a variant of timed Petri nets,the standard model of body iron homeostasis is enriched by including the durations of the pertinent biochemical reactions.A Petri-net variant in which, at each node,a time interval is specified is used in order to describe the time lag of the commencement of conditions that must be fulfilled before a biochemical reaction can start.Due to critical changes in the environment,switches can occur in metabolic networks that lead to systems exhibiting simultaneously discrete and continuous dynamics.Hybrid systems represent this accurately.The paper Modeling and simulation of metabolic networks for estimation of biomass-accumulation parameters by Uˇg ur Kaplan,Metin Türkay,Bülent Karasözen,and Lorenz Biegler develops a hybrid system to simulate cell-metabolism dynamics that includes the effects of extra-cellular stresses on metabolic responses.Path-finding approaches to metabolic-pathway analysis adopt a graph-theoretical approach to determine the reactions that an organism might use to transform a source compound into a target compound.In the contribution Path-finding approaches and metabolic pathways,Francisco J.Planes and John E.Beasley examine the effectiveness of using compound-node connectivities in a path-finding approach.An approach to path finding based on integer programming is also presented. Existing literature is reviewed.This paper is well illustrated and provides many examples as well as,as an extra service,some supplementary information.In A new constraint-based description of the steady-state flux cone of metabolic networks,Abdelhalim Larhlimi and Alexander Bockmayr present a new constraint-based approach to metabolic-pathway analysis.Based on sets of non-negativity constraints,it uses a description of the set of all possible flux distributions over a metabolic network at a steady state in terms of the steady-state flux cone.The constraints can be identified with irreversible reactions and,thus,allow a direct interpretation.The resulting description of the flux cone is minimal and unique.Furthermore,it satisfies a simplicity condition similar to the one for elementary flux modes.Most biological networks share some properties like being,e.g.,‘‘scale free’’.Etienne Birmeléproposes a new random-graph model in his contribution A scale-free graph model based on bipartite graphs that can be interpreted in terms of metabolic networks,and exhibits this specific feature.$Dedicated to our dear teacher and friend Prof.Dr.Peter Ladislaw Hammer(1936–2006).0166-218X/$–see front matter©2009Elsevier B.V.All rights reserved.doi:10.1016/j.dam.2009.01.0212218Preface/Discrete Applied Mathematics157(2009)2217–2220Differential equations have been established to quantitatively model the dynamic behaviour of regulatory networks representing interactions between cell components.In the paper Inference of an oscillating model for the yeast cell cycle, Nicole Radde and Lars Kaderali study differential equations within a Bayesian setting.First,an oscillating core network is learned that is to be extended,in a second step,using‘‘Bayesian’’methodology.A specifically designed hierarchical prior distribution over interaction strengths prevents overfitting and drives the solutions to sparse networks.An application to a real-world data set is provided,and its dynamical behaviour is reconstructed.The contribution An introduction to the perplex number system by Jerry L.R.Chandler derives from his approach to theoretical chemistry,and provides a universal source of diagrams.The perplex number system,a new logic for describing relationships between concrete objects and processes,provides in particular an exact notation for chemistry without invoking either chemical or‘‘alchemical’’symbols.Practical applications to concrete compounds(e.g.,isomers of ethanol and dimethyl ether)are given.In conjunction with the real number system,the relations between perplex numbers and scientific theories of concrete systems(e.g.,intermolecular dynamics,molecular biology,and individual medicine)are described.Since exact determination of haplotype blocks is usually impossible,a method is desired which can account for recombinations,especially,via phylogenetic networks or a simplified version.In their work Haplotype inferring via galled-tree networks using a hypergraph-covering problem for special genotype matrices,Arvind Gupta,Ján Maňuch,Ladislav Stacho, and Xiaohong Zhao reduce the problem via galled-tree networks to a hypergraph-covering problem for genotype matrices satisfying a certain combinatorial condition.Experiments on real data show that this condition is mostly satisfied when the minor alleles(per SNP)reach at least30%.Recently the Quartet-Net or,for short,‘‘QNet’’method was introduced by Stefan Grünewald et al.as a method for computing phylogenetic split networks from a collection of weighted quartet trees.Here,Stefan Grünewald,Vincent Moulton,and Andreas Spillner show that QNet is a‘‘consistent’’method.This key property of QNet does not only guarantee to produce a tree if the input corresponds to a tree—and an outer-labeled planar split network if the input corresponds to such a network;the proof given in their contribution Consistency of the QNet algorithm for generating planar split networks from weighted quartets also provides the main guiding principle for the design of the method.Kangal and Akbash dogs are the two well-known shepherd dog breeds in Turkey.In the article The genetic relationship between Kangal,Akbash,and other dog populations,Evren Koban,Çigdem Gökçek Saraç,Sinan Can Açan,Peter Savolainen, andİnci Togan present a comparative examination by mitochondrial DNA control region,using a consensus neighbour-joining tree with bootstrapping which is constructed from pairwise FST values between populations.This study indicates that Kangal and Akbash dogs belong to different branches of the tree,i.e.,they might have descended maternally from rather different origins created by an early branching event in the history of the domestic dogs of Eurasia.In their paper The Asian contribution to the Turkish population with respect to the Balkans:Y-chromosome perspective,Ceren Caner Berkman and inci Togan investigate historical migrations from Asia using computational approaches.The admixture method of Chikhi et al.was used to estimate the male genetic contribution of Central Asia to hybrids.The authors observed that the male contribution from Central Asia to the Turkish population with reference to the Balkans was13%.Comparison of the admixture estimate for Turkey with those of neighboring populations indicated that the Central Asian contribution was lowest in Turkey.Split-decomposition theory deals with relations between real-valued split systems and metrics.In his work Split decomposition over an Abelian group Part2:Group-valued split systems with weakly compatible support,Andreas Dress uses a general conceptual framework to study these relations from an essentially algebraic point of view.He establishes the principal results of split-decomposition theory regarding split systems with weakly compatible support within this new algebraic framework.This study contributes to computational biology by analyzing the conceptual mathematical foundations of a tool widely used in phylogenetic analysis and studies of bio-diversity.The contribution Phylogenetic graph models beyond trees of Ulrik Brandes and Sabine Cornelsen deals with methods for phylogenetic analysis,i.e.,the study of kinship relationships between species.The authors demonstrate that the phylogenetic tree model can be generalized to a cactus(i.e.,a tree all of whose2-connected components are cycles)without losing computational efficiency.A cactus can represent a quadratic rather than a linear number of splits in linear space.They show how to decide in linear time whether a set of splits can be accommodated by a cactus model and,in that case,how to construct it within the same time bounds.Finally,the authors briefly discuss further generalizations of tree models.In their paper Whole-genome prokaryotic clustering based on gene lengths,Alexander Bolshoy and Zeev Volkovich present a novel method of taxonomic analysis constructed on the basis of gene content and lengths of orthologous genes of 66completely sequenced genomes of unicellular organisms.They cluster given input data using an application of the information-bottleneck method for unsupervised clustering.This approach is not a regular distance-based method and, thus,differs from other recently published whole-genome-based clustering techniques.The results correlate well with the standard‘‘tree of life’’.For characterization of prokaryotic genomes we used clustering methods based on mean DNA curvature distributions in coding and noncoding regions.In their article Prokaryote clustering based on DNA curvature distributions,due to the extensive amount of data Limor Kozobay-Avraham,Sergey Hosida,Zeev Volkovich,and Alexander Bolshoy were able to define the external and internal factors influencing the curvature distribution in promoter and terminator regions.Prokaryotes grow in the wide temperature range from4◦C to100◦C.Each type of bacteria has an optimal temperature for growth.They found very strong correlation between arrangements of prokaryotes according to the growth temperature and clustering based on curvature excess in promoter and terminator regions.They found also that the main internal factors influencingPreface/Discrete Applied Mathematics157(2009)2217–22202219 the curvature excess are genome size and A+T composition.Two clustering methods,K-means and PAM,were applied and produced very similar clusterings that reflect the aforementioned genomic attributes and environmental conditions of the species’habitat.The paper Pattern analysis for the prediction of fungal pro-peptide cleavage sites by SüreyyaÖzöˇgür Ayzüz,John Shawe-Taylor,Gerhard-Wilhelm Weber,and Zümrüt B.Ögel applies support-vector machines to predict the pro-peptide cleavage site of fungal extra-cellular proteins displaying mostly a monobasic or dibasic processing site.A specific kernel is expressed as an application of the Gaussian kernel via feature spaces.The novel approach simultaneously performs model selection, tests the accuracy,and computes confidence levels.The results are found to be accurate and compared with the ones provided by a server.Preetam Ghosh,Samik Ghosh,Kalyan Basu,and Sajal Das adopt an‘‘in silico’’stochastic-event-based simulation methodology to determine the temporal dynamics of different molecules.In their paper Parametric modeling of protein–DNA binding kinetics:A discrete event-based simulation approach,they present a parametric model for predicting the execution time of protein–DNA binding.It considers the actual binding mechanism along with some approximated protein-and DNA-structural information using a collision-theory-based approach incorporating important biological parameters and functions into the consideration.Murat Ali Bayır,Tacettin Doˇg acan Güney,and Tolga Can propose a novel technique in their paper Integration of topological measures for eliminating non-specific interactions in protein interaction networks for removing non-specific interactions in a large-scale protein–protein interaction network.After transforming the interaction network into a line graph,they compute betweenness and other clustering coefficients for all the edges in the network.The authors use confidence estimates and validate their method by comparing the results of a test case relating to the detection of a molecular complex with reality.The article Graph spectra as a systematic tool in computational biology by Anirban Banarjee and Jürgen Jost deals with the obviously important question of how biological content can be extracted from the graphs to which biological data are often reduced.From the spectrum of the graph’s Laplacian that yields an essentially complete qualitative characterization of a graph,a spectral density plot is derived that can easily be represented graphically and,therefore,analyzed visually and compared for different classes of networks.The authors apply this method to the study of protein–protein interaction and other biological and infrastructural networks.It is detected that specific such classes of networks exhibit common features in their spectral plots that readily distinguish them from other classes.This represents a valuable complement to the currently fashionable search for universal properties that hold across networks emanating from many different contexts.Konstantin Klemm and Peter F.Stadler’s Note on fundamental,nonfundamental,and robust cycle bases investigates the mutual relationships between various classes of cycle bases in a network that have been studied in the literature.The authors show for instance that strictly fundamental bases are not necessarily cyclically robust;and that,conversely, cyclically robust bases are not necessarily fundamental.The contribution focuses on cyclically robust cycle bases whose existence for arbitrary graphs remains open despite their practical use for generating all cycles of a given2-connected graph. It presents also a class of cubic graphs for which cyclically robust bases can be constructed explicitly.Understanding the interplay and function of a system’s components also requires the study of the system’s functional response to controlled experimental perturbations.For biological systems,it is problematic with an experimental design to aim at a complete identification of the system’s mechanisms.In his contribution A refinement of the common-cause principle,Nihat Ay employs graph theory and studies the interplay between stochastic dependence and causal relations within Bayesian networks and information theory.Applying a causal information-flow measure,he provides a quantitative refinement of Reichenbach’s common-cause principle.Based on observing an appropriate collection of nodes of the network, this refinement allows one to infer a hitherto unknown lower bound for information flows within the network.In their article Discovering cis-regulatory modules by optimizing barbecues,Axel Mosig,Türker Bıyıkoˇg lu,Sonja J.Prohaska, and Peter F.Stadler ask for simultaneously stabbing a maximum number of differently coloured intervals from K arrangements of coloured intervals.A decision version of this best barbecue problem is shown to be NP-complete.Because of the relevance for complex regulatory networks on gene expression in eukaryotic cells,they propose algorithmic variations that are suitable for the analysis of real data sets comprising either many sequences or many binding sites.The optimization problem studied generalizes frequent itemset mining.The contribution A mathematical program to refine gene regulatory networks by Guglielmo Lulli and Martin Romauch proposes a methodology for making sense of large,multiple time-series data sets arising in expression analysis.It introduces a mathematical model for producing a reduced and coherent regulatory system,provided a putative regulatory network is given.Two equivalent formulations of the problem are given,and NP-completeness is established.For solving large-scale instances,the authors implemented an ant-colony optimization procedure.The proposed algorithm is validated by a computational analysis on randomly generated test instances.The practicability of the proposed methodology is also shown using real data for Saccharomyces cerevisiae.Jutta Gebert,Nicole Radde,Ulrich Faigle,Julia Strösser,and Andreas Burkovski aim in their paper Modelling and simulation of nitrogen regulation in Corynebacterium glutamicum at understanding and predicting the interactions of macromolecules inside the cell.It sets up a theoretical model for biochemical networks,and introduces a general method for parameter estimation,applicable in the case of very short time series.This approach is applied to a special system concerning nitrogen uptake.The equations are set up for its main components,the corresponding optimization problem is formulated and solved, and simulations are carried out.2220Preface/Discrete Applied Mathematics157(2009)2217–2220Gerhard-Wilhelm Weber,Ömür Uˇg ur,Pakize Taylan,and Aysun Tezel model and predict gene-expression patterns incorporating a rigorous treatment of environmental aspects,and aspects of errors and uncertainty.For this purpose,they employ Chebyshev approximation and generalized semi-infinite optimization in their paper On optimization,dynamics and uncertainty:A tutorial for gene–environment networks.Then,time-discretized dynamical systems are studied,the region of parametric stability is detected by a combinatorial algorithm and,then,the topological landscape of gene–environment networks is analyzed in terms of its‘‘structural stability’’.We are convinced that all papers selected for this special issue constitute valuable contributions to many different areas in computational biology,employing methods from discrete mathematics and related fields.We again thank all colleagues who have participated in this exciting endeavor with care,foresight,and vision,for their highly appreciated help.Guest editorsAndreas DressBülent KarasözenPeter F.StadlerGerhard-Wilhelm Weber125July2008Available online29March2009 1Assistant to the guest editors:Mrs.Cand.MSc.Bengisen Pekmen(Institute of Applied Mathematics,METU,Ankara).。
Standard Model
1 Historical background
The Standard Model of elementary particles (more schematic depiction), with the three generations of matter, gauge bosons in the fourth column, and the Higgs boson in the fifth.
3.1 Fermions
quarks being very strongly bound to one another, forming color-neutral composite particles (hadrons) containing either a quark and an antiquark (mesons) or three quarks (baryons). The familiar proton and the neutron are the two baryons having the smallest mass. Quarks also carry electric charge and weak isospin. Hence they interact with other fermions both electromagnetically and via the weak interaction.
Standard Model
This article is about the Standard Model of particle physics. For other uses, see Standard model (disambiguation). This article is a non-mathematical general overview of the Standard Model. For a mathematical description, see the article Standard Model (mathematical formulation). For the Standard Model of Big Bang cosmology, Lambda-CDM model. The Standard Model of particle physics is a theory con-
Particle-Imaging Techniques For Experimental Fluid Mechanics
I
Speckle Patterns
Particulate Markers
(N.»l)
I I
LSV
Particle Images
I
I I
I
I II
I
Fluorescent
Molecular Markers
Photochromic
I
I
I
I
(N.«l)
Particle-Image Velocimetry
A technique that uses particles and their images falls into the category commonly known as particle-image ve/ocimetry, or PI V, which is the principal subject of this article. Before comparing the characteristics of PIV with the other methods displayed in Figure I, it is helpful to examine
PIV
I
NI» l
High Image Density PIV
Low Image Density PlY PlY
NI«l
Figure 1
Particle-image velocimetry and other forms of pulsed-light velocimetry.
PARTICLE-IMAGING TECHNIQUES
where ilx is the displacement of a marker, located at x at time t, over a short time interval Llt separating observations of the marker images. The particles are usually solids in gases or liquids but can also be gaseous bubbles in liquids or liquid droplets in gases or immiscible liquids. Other types of markers include (a) patches of molecules that are activated by laser beams, causing them either to fluoresce (Gharib et al 1985), or to change their optical density by photochromic chemical reactions (Popovich & Hummel 1967, Ricka 1987), and (b) speckle patterns caused by illumi nating groups of particles with coherent light. Regardless of the marker type, locations at various instants are recorded optically by pulses of light that freeze the marker images on an optical recording medium such as a photographic film, a video array detector, or a holographic film. Since these methods share many similarities, it is useful to group them under the single topic of pulsed-light velocimetry, or PLV. The various P LV techniques are organized in Figure I.
Particle size analysis utilizing polarization inte
专利名称：Particle size analysis utilizing polarization intensity differential scattering发明人：Steven E. Bott,W. Howard Hart申请号：US07/575797申请日：19900831公开号：US05104221A公开日：19920414专利内容由知识产权出版社提供摘要：Two arrangements are disclosed to provide high resolution measurement of sub-micrometer and micrometer particle size distributions. In a first arrangement, scattered light is measured over a wide range of scattering angles. At the same time, light scattered at low scattering angles is measured with high angular resolution. In the second arrangment, an improved Polarization Intensity Differential Scattering (PIDS) measurement is made possible by providing an interrogating light beam of selected wavelength including a first component having a linear polarization plane and a second component having a differential linear polarization plane, wherein the linear polarizations of the components are orthogonal. Photodetecting arrays in one or more scattering planes detect light scattered by the particles at least at two scattering angles.申请人：COULTER ELECTRONICS OF NEW ENGLAND, INC.代理机构：Lahive & Cockfield更多信息请下载全文后查看。
轻暗物质SZ效应-课件
第十届全国粒子物理学术会议
7
三类SZ效应：
热电子的随机运动——热SZ效应 电子集体(星系团)的整体运动——运动学SZ效应 其它非热电子集体——非热SZ效应
SZ效应
第十届全国粒子物理学术会议
8
暗物质简述 Sunyaev-Zel’dovich (SZ)效应 ~MeV暗物质所致SZ效应 总结和讨论
第十届全国粒子物理学术会议
9
MeV暗物质，被引入解释银心511keV发射 MeV暗物质SZ效应
→e+e- , dN/dE=(E-m) 传播过程：扩散，能量损失
<v>(1MeV/m)2~10-30cm3/s PRL, 2004, 92, 101301
A&A, 2006, 455, 21
湮灭电子密度平方依赖于暗物质 数密度分布：低质量暗物质粒子 能产生更多正负电子
在频率217GHz处，热SZ效应为零 M=10MeV，T~3K M=100MeV，T~30K
第十届全国粒子物理学术会议
12
暗物质简述 Sunyaev-Zel’dovich (SZ)效应 ~MeV暗物质所致SZ效应 总结和讨论
第十届全国粒子物理学术会议
13
总结
暗物质自湮灭产生正负电子对，正负电子 和CMB光子相互作用导致SZ效应
第十届全国粒子物理学术会议
5
暗物质简述 Sunyaev-Zel’dovich (SZ)效应 ~MeV暗物质所致SZ效应 总结和讨论
第十届全国粒子物理学术会议
6
Sunyaev & Zel’dovich
SZ效应
CMB光子与宇宙中高能 电子逆康普顿散射，低 能光子变少，高能光子 增多，导致CMB黑体谱 的畸变。
Numerical Predication of Cracking Reaction of Particle Clusters in
Chinese Journal of Chemical Engineering, 16(5) 670ü678 (2008)Numerical Predication of Cracking Reaction of Particle Clusters in Fluid Catalytic Cracking Riser Reactors*WANG Shuyan ( )1,**, LU Huilin ( )1, GAO Jinsen ( )2, XU Chunming ( ҝ )2 and SUN Dan ( ӥ)11 School of Energy Science and Engineering, Harbin Institute of Technology, Harbin 150001, China2 State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Beijing 102249, ChinaAbstract Behavior of catalytic cracking reactions of particle cluster in fluid catalytic cracking (FCC) riser reac-tors was numerically analyzed using a four-lump mathematical model. Effects of the cluster porosity, inlet gas ve-locity and temperature, and coke deposition on cracking reactions of the cluster were investigated. Distributions of temperature, gases, and gasoline from both catalyst particle cluster and an isolated catalyst particle are presented.The reaction rates from vacuum gas oil (VGO) to gasoline, gas and coke of individual particle in the cluster are higher than those of the isolated particle, but it reverses for the reaction rates from gasoline to gas and coke. Less gasoline is produced by particle clustering. Simulated results show that the produced mass fluxes of gas and gasoline increase with the operating temperature and molar concentration of VGO, and decrease due to the formation of coke.Keywords cluster, fluid catalytic cracking, numerical simulation, riser1 INTRODUCTIONFluid catalytic cracking (FCC) is considered to be one of the most important petroleum refining proc-esses. The fluid catalytic cracking involves the con-version of heavy oil feedstock into gasoline and other valuable products [1]. The FCC unit comprises mainly of a riser and a regenerator. The vacuum gas oil (VGO) is dispersed into the riser bottom in the form of drops through a feed nozzle system. The VGO drops contact hot regenerated catalyst particles from the regenerator and get vaporized. The vapor entrains the catalyst par-ticles and the liquid drops getting cracked on the cata-lyst surface along the riser. In this process, the catalyst is progressively deactivated due to the deposition of coke formed during cracking reactions on its surface. The deactivated catalyst leaving the riser top is trans-ferred to the regenerator where its activity is restored by burning the coke. Therefore, hydrodynamics of the catalyst particle in the riser controls the performance of the FCC unit.It is generally agreed that the suspended particles may partially form denser clusters in the riser. These clusters move upward through the center of the riser and fall down along the riser wall [2]. Flow charac-terization of clusters has been experimentally investi-gated in the circulating fluidized beds (CFBs) [3, 4]. The particle clustering leads to inefficient gas-solids contact, and therefore affects on CFB performance. Hence, in-depth knowledge of the cluster behavior is essential for designing and operating the FCC units.Many numerical simulations have been per-formed to obtain the performance of the FCC unit based on catalyst reaction kinetic models [5, 6]. Most of them represents oil components in a few lumps, including the three-lump cracking model [7] to study gasoline production in a FCC unit and the four-lump model [8, 9] to describe the catalytic cracking of gaso-line, the five-lump model to consider the heavy frac-tion and with the gas oil splitting into paraffins, naphthenes, and aromatics [10], and the more detailed ten-lump kinetic model to take into account for dif-ferent feed properties in addition to the boiling point range [11]. Flow and cracking reactions have been predicted by using a three-dimensional computational fluid dynamics (CFD) models for FCC riser with Eul-erian approach and reactions with lumping models. Gupta and Subbarao [12] developed a three-phase model for FCC riser to analyze the effect of feed at-omization on conversion. Obviously, the number of lumps of the proposed models for catalytic cracking reactions may be increased to obtain more detailed descriptions of catalytic cracking reactions and prod-uct distributions [13, 14]. However, the cracking reac-tion models mentioned above have not considered the particle clustering effect on the cracking reactions as well as the flow behavior in the FCC riser reactors.Experiments suggested that the flow behavior of particle clusters be significantly different from single particles in the same flow stream. The clustering plays an important role on performance of CFB risers. It is, therefore, desirable to know characteristic behavior of particles in the clusters compared with that of dis-persed particles. The effects of catalyst particle clus-ters on cracking reaction in a FCC riser reactor is here numerically analyzed. Chemical reactions involve heterogeneous and homogeneous reactions among gas, gasoline and coke. Numerical results provide distribu-tions of temperature, gas and gasoline in the cluster. Effects of inlet temperature and velocity of mixture gases, cluster porosity, and coke formation on the mass fluxes of gas and gasoline were also investigated.2 GOVERNING EQUATIONSFor simplicity, the cluster is assumed to be spherical with a diameter of Dc containing 65 spheri-cal particles as shown in Fig. 1. The particles in theReceived 2007-10-09, accepted 2008-08-06.* Supported by the National Natural Science Foundation of China (50776023) and NSFC-Petro China (20490200). ** To whom correspondence should be addressed. E-mail: wangshuyan@Chin. J. Chem. Eng., Vol. 16, No. 5, October 2008 671cluster are regularly arranged and assumed to be sta-tionary during simulations. The cluster porosity is de-fined as333g c p c /D nd D H . With the particle di-ameter of 100 μm and the cluster size of 688 μm con-taining 65 particles in the cluster, the corresponding porosity is 0.8. Following assumptions are made: (1) The catalyst particle diameter and shape remains con-stant during the reactions; (2) Radiation heat transfer between gases and particles is neglected because of the small temperature difference between gases and particles, and radiation heat transfer between particles and the riser walls is not considered; (3) The instanta-neous vaporization of feeding at the riser entry is as-sumed and the post combustion reaction is not con-sidered; (4) Catalyst particles are very small (and its thermal conductivity is very high), therefore, tem-perature difference between its surface and center may be neglected for individual particles.The vaporized vacuum gas oil (VGO) diffuses into particulate phase and is cracked at the catalyst sur-face. While produced gases diffuse back to gas phase, and the produced coke is simultaneously deposited onto the catalyst surface. The four-lump model (Fig. 2) for cracking reactions kinetics is used in present analysis. The four lumps are feed (VGO), gasoline, gas and coke. The primary cracking of gas-oil is as-sumed to be first order kinetics. The turbulent flow field is computed using a k -İ turbulent model. The conservation equations of mass, momentum, gas speciesand energy in Eulerian coordinates are given as follows.Figure 2 4-lump kinetic model2.1 Gas phase continuity equationThe mass conservation equation of gas can be written as [15, 16]:0j ju t x U U w w w w , 14j (1) 2.2 Gas phase momentum equationThe conservation of momentum for the gas phase is expressed as follows:g g g j i j i j i j i i jj u p u u u t x x x x u g x x U U P P U w §·w w w w¨¸w w w w w ©¹§·w w ¨¸¨¸w w ©¹(2) where g l t P P P is the effective gas viscosity, and the turbulent viscosity 2t /C k P P U H is determined from a k -İ turbulent model.2.3 Gas phase turbulent k -İ modelHere, we employ the standard k -İ turbulence model for simulating turbulent flow in the riser reactor. The standard k -İ turbulence model gives the following turbulent kinetic energy equation [16]:t g R ()j k j jk j k k u k G G t x x x P U U U H V §·w w ww¨¸¨¸w w w w ©¹(3)and the turbulent kinetic energy dissipation rate equationt 121R ()j j j j k u t x x x C G C C G k kP H UH U H V HHUH H §·w w w w¨¸¨¸w w w w ©¹(4)where G k is the production term. The empirical con-stants, C P , V H , C 1 and C 2, are listed in Table 1.2.4 Energy conservation of gas phaseFor the energy balance equation of the gas phaseis written as follows [15, 16]:g g g g g t g g h j ji i i jj C T u C T t x C T H R x x Q O V w w w w ªºw §·w'«»¨¸w w «»©¹¬¼¦(5)(a) Front view (b) Cross-sectional view (0z )Figure 1 The cluster studied in the simulationsChin. J. Chem. Eng., Vol. 16, No. 5, October 2008 672where R i is the reaction rates of route A, B, C, D and E.The heat capacity and the thermal conductivity of the mixture are calculated by the mixing-law:g g i i iC Y C ¦ (6)g g i i iY O O ¦(7) where C g i and Ȝg i are the heat capacity and the thermal conductivity of species i respectively. The heat capac-ity and the thermal conductivity of gas and gasoline were estimated respectively using a modified Lee-Kesler correlation [17] in which the pressure ef-fect is neglected since the typical FCC processes are operated under a relatively low pressure. 2.5 Conservation equation of gas speciesThe mass balance for a gas species k is:g g t Y k j k jk k k i i i jj Y u Y t x Y D M R x x U U P U V w ww w ªº§·w w «»¨¸w w «»©¹¬¼¦ (8) where ıY is the turbulent Schmidt number with ıY ˙0.7. The last term on the right hand side of Eq. (8) is the mass source from the reactions.2.6 Cracking reactions of single catalyst particleFour-lump reaction kinetic models were consid-ered in the present work to represent gas phase cata-lytic cracking reactions. The reaction schemes for these kinetic models are shown in Fig. 2. The four lumps are: VGO (feedstock), gasoline, gas and coke. VGO is cracked to gasoline, gas and coke [18]. Therate of consumption of reactant k per unit catalyst volume can be expressed as [19]:0nk k rk k k C R K C C M §·¨¸©¹ (9) where C k is the concentration of gas component, and C k 0 is the initial concentration of pure component k . The value of n is set to be unit for VGO cracking. Thisimplies that VGO evaporated at the beginning of the cracking reaction is easier to crack than VGO evapo-rated later since heavier components take longer to evaporate. For all other reactions the value of n is set to be zero. The temperature dependence of kineticparameters appearing in Eq. (9) is described by the Arrhenius expression:g exp /rk k k K A E RT (10)The FCC catalyst can be deactivated due to thecoke formation [18, 19]. The coke deposited on the surface of the spent catalyst affects coke-oxidation kinetics. The parameter, M , in Eq. (9) represents the activity factor. Non-selective deactivation of catalystis assumed. The activity factor Mis related to the coke deposition on the catalyst as [18, 19]:cc c c 1exp B B A C M(11)where C c is the coke concentration (mass percentage). The values for deactivation constants A c and B c were taken as 4.29 and 10.4, respectively. The kinetic con-stants for the four lump model used in present simula-tions are listed in Table 2 [20, 21].2.7 Boundary conditions and simulation proceduresInitially, both gas and particle temperatures were set to be 991 K. At the inlet, velocity, gas composition,Table 1 Parameters used in the simulationsinlet gas velocity 1.0 m·s ˉ1diameter of particle 100 m inlet temperature 991 K size of cluster 688 m inlet mass fraction of VGO 0.27 number of particles 65 inlet mass fraction of steam 0.73 porosity of cluster0.8 particle specific heat 1000 J·kg ˉ1·K ˉ1particle density 1730 kg·m ˉ3particle thermal conductivity 0.0454 W·m ˉ1·K ˉ1temperature of particle 991 Kmolecular mass of VGO 400 kg·kmol ˉ1viscosity of VGO 5.0×10ˉ5 kg·m ˉ1·s ˉ1 molecular mass of gas 50 kg·kmol ˉ1viscosity of gas 1.66×10ˉ5 kg·m ˉ1·s ˉ1molecular mass of gasoline 100 kg·kmol ˉ1viscosity of gasoline 1.66x10ˉ5 kg·m ˉ1·s ˉ1molecular mass of steam 18 kg·kmol ˉ1viscosity of steam 2.0x10ˉ5 kg·m ˉ1·s ˉ1molecular mass of coke 400 kg·kmol ˉ1viscosity of coke1.66×10ˉ5 kg·m ˉ1·s ˉ1diffusion coefficient of VGO 8.8×10ˉ6 m 2·s ˉ1diffusion coefficient of gasoline 1.7×10ˉ5 m 2·s ˉ1diffusion coefficient of gas2.064×10ˉ5 m 2·s ˉ1diffusion coefficient of steam2.178×10ˉ5 m 2·s ˉ1empirical constant C 1 1.44 empirical C 2 1.92 empirical constant C ȝ 0.09 empirical ık 1.0 empirical constant ıİ 1.3empirical ıh 1.0Chin. J. Chem. Eng., Vol. 16, No. 5, October 2008 673temperature, turbulent kinetic energy and energy dis-passion are specified:g,g y u u ; g,g,0x z u u ; ,0i i Y Y ; g g,0T T ;g 0.004k u ; 3/2m /k H O (12)where the turbulent mixing length m O is taken as m c 2D O . At the outlet, the boundary conditions are g,g,g,0y x z u u u y x z w w w w w w ; 0i Yyw w ; g 0T y w w ; 0kyw w ; 0y H w w (13) On the particle surface, no-slip boundary condi-tion for gas flow was assumed:0u k H (14)For an isothermal catalytic particle, the tempera-ture at the surface of particle is given, and the mass fractions of gases, gasoline and feedstock are0i Y (15)The governing Eqs. (1) (8) coupled with reaction kinetics and boundary conditions are solved using a finite volume method [22]. The modeled part of the three-dimensional riser and its dimensions are shown in Fig. 1. The cluster is assumed to be located in the center of the computational domain. Each particle in the cluster is labeled. Because the cluster size is so small compared with the riser size, the riser wall effect on the gas flow around the cluster can be ignored.The three-dimensional conjugate flow, heat trans-fer and reactions were solved using the CFD software FLUENT 6.2. Reactions and gases properties were incorporated to the solver code through user defined functions (UDF). Appropriate mixing rules were ap-plied to the model for making all variables composition- dependent. The SIMPLE algorithm was used to couple the pressure and the velocities with the second order upwind method used in the momentum equations and for the energy equation. Fresh gases flow into the cluster and react on the surface of catalyst particles. A set of simulation needs about 24 h of CPU time on PC computer (80 GB hard disk, 128 Mb Ram and of 600 MHz CPU).2.8 Computational domain and gridThe conservation equations of momentum, en-ergy and gas species have been solved numericallyusing the segregated solver module of FLUENT 6.2. Using GAMBIT, a structured grid of triangular cells was created. The unstructured mesh used about 300000 elements. Each particle had 200 300 nodes with 700 1000 elements. The mesh was continually refined up to the point beyond which any further re-finement changed the solution by less than 1.0% at the expense of an enormous increase in CPU time. The effect of number of nodes on the surface of a particle on the drag coefficient is shown in Table 3. Therefore, the results presented herein are believed to be free from domain and grid effects as the number of nodes on a particle is larger than 200.Table 3 Comparison of drag coefficient for anisolated particlePresent simulations (number of nodes on a particle)Re 100 200 300Johnson and Patel [23]10 5.445 4.013 3.986 3.9 20 3.673 2.587 2.515 2.50 50 2.036 1.583 1.534 1.50 1001.546 1.112 1.068 1.053 COMPUTED RESULTS AND ANALYSIS 3.1 Base case simulationsIn the base case simulation, the inlet gas velocity,inlet temperature and porosity were set to be 1.0 m·s ˉ1, 991 K and 0.8, respectively. The terminal velocity ofparticle is about 0.39 m·s ˉ1.In order to validate the simulation, a single sphere suspended in a computational domain was cre-ated. In this work, gas flow past an unconfined sphere which is similar to a particle located in the center of the cluster. The values of drag coefficient are calcu-lated for a range of particle Reynolds number. Table 3 shows the comparison between the present results with the literature values. Three different number of nodes on a particle are used. Present simulations are in agreement with the data in Ref. [23] at the nodes number of 200 300 on a particle.Figure 3 shows the distributions of mass fraction and temperature of gases through the catalyst particle cluster (0Y , 0Z ). The Reynolds number is 4145 based on cluster diameter. The calculated mean gasvelocity inside the cluster is about 0.057 m·s ˉ1 at theinlet gas velocity and porosity of cluster of 1.0 m·s ˉ1 and 0.8. The Reynolds number is 34.4 based on particleTable 2 Kinetics constants used in present simulationsCracking reactionPre-exponential factor A j /m 3·m ˉ3·s ˉ1 Activation energy E j /J·kmol ˉ1ǻH r /J·kg ˉ1route A: AVGO gasoline Ko 3.5890×106 6.836×107 195 route B: BVGO gas Ko 2.5410×107 8.936×107 670 route C: CVGO coke Ko 6.7910×107 6.468×107 745 route D: Dgasoline gas Ko 8.8572×102 5.280×107 512.5 route E: Egasoline coke Ko 5.3198×107 1.1565×108 550Chin. J. Chem. Eng., Vol. 16, No. 5, October 2008674diameter which indicates laminar flow inside the clus-ter. Flow behavior of gases around a particle inside thecluster is different from an isolated particle in the stream. This indicates that the flow around a cluster is similar to the flow around a porosity body. The mass fractions and temperature of gases are spatially varied continuously from the inside to outside of a cluster.Various rates for particle in the cluster and for the isolated particle are shown in Fig. 4. It is found that the reaction rates from VGO to gasoline, gas and coke are higher in the front than those in the back of the cluster, while the reaction rates from gasoline to gas and coke are lower in the front than those in the back of the cluster. For the isolated particle, the reaction rates from VGO to gasoline, gas and coke are higher, and the rate from gasoline to gas and coke is lower comparing with individual particle in the cluster. Fig. 5 shows the produced gas and gasoline molar dis-tributions from each particle in the cluster and from the isolated particle. It is found that the produced gas and gasoline from individual particle in the cluster are more than those from the isolated particle due to the accumulation effect along flow direction. There is avariation of the produced gas and gasoline from parti-cle to particle in the cluster. 3.2 Effect of porosity of clusterThe ratio of gas flux through the cluster to inlet gas flux with the same flowing cross-sectional area of the cluster is shown in Fig. 6 as a function of cluster porosity. With cluster diameters of 638, 757, 866 and 1090 μm, the corresponding cluster porosities are 0.75, 0.85, 0.9 and 0.95, respectively. With the increase of concentration of particles in the cluster, the ratio of gas flux is reduced due to the high flow resistance inside the cluster. Simulated results show that the variation of gas temperature inside the cluster is notobvious with the changes of cluster porosity.Figure 6 Profile of ratio of gas flux through cluster to inlet air flux and averaged gas temperature (u g ˙1 m·s ˉ1, T ˙991 K, mass fraction of VGO ˙0.27, mass fraction of steam ˙0.73, number of particles in the cluster ˙65)With the increase of cluster porosity, more VGO can pass through the cluster and results in the cracking reaction rate increment, as shown in Fig. 7. The reac-tion rates from VGO to gasoline, gas and coke are increased, while the reaction rates from gasoline to gas and coke are decreased with the increase of cluster porosity. The reaction rate of gasoline to gas is almost one magnitude order less than that of VGO to gasoline.These results in the more gasoline and less gas produced,Figure 3 Profile of mass fractions and temperature of gases through the cluster (0Y ,0Z )[u g ˙1 m·s ˉ1, İg ˙0.8, T ˙991 K, mass fraction of steam ˙0.73, mass fraction of VGO ˙0.27 (or concentration of VGO ˙0.2 mol·m ˉ3)]Figure 4 Profile of reaction rates of particles in the clus-ter and an isolated particle (u g ˙1 m·s ˉ1, T ˙991 K, mass fraction of VGO ˙0.27, mass fraction of steam ˙0.73) 1ü11particle in the cluster; 12 isolated particleFigure 5 Distribution of molar fraction of gas and gaso-line from particles in the cluster and isolated particle (u g ˙1.0 m·s ˉ1; İg ˙0.8; T ˙991 K; mass fraction of VGO ˙0.27, mass fraction of steam ˙0.73)1ü11 particles in cluster; 12 isolated particle; gas; gasolineChin. J. Chem. Eng., Vol. 16, No. 5, October 2008 675as shown in Fig. 8. More gasoline and less gas will beproduced with the increase of particle concentration of the cluster. Hence, the particle clustering on the cracking reactions in the FCC riser is obvious.Figure 8 Distribution of mass fluxes of gas and gasolinefrom isolated particle and particles in the cluster [Both num-bers of isolated particles and particles in the cluster are 65, u g ˙1 m·s ˉ1; T ˙991 K; mass fraction of VGO ˙0.27 (or concen-tration of VGO ˙0.2 mol·m ˉ3), mass fraction of steam ˙0.73] Ƶ gas; ƻ gasoline3.3 Effect of inlet gas velocityFigure 9 shows the reaction rates of the cluster and the isolated particle as a function of inlet gas ve-locity. Roughly, the reaction rates from VGO to gaso-line, gas, and coke increase, while the reaction rates from gasoline to gas and coke decrease with the in-crease of inlet gas velocity. For the isolated particle, the effects of inlet gas velocity on the reaction rates are not noticeable.The flow rates of gasoline and gas can be calcu-lated by summing the product of density and mass fraction and velocity with the product of the facet area vector and the facet velocity vector at each grid. Fig. 10 shows the yields of gas and gasoline from the cluster and from the isolated particle as a function of inlet gas velocity. Within the range of gas inlet veloci-ties of 0.4 m·s ˉ1 to 1.6 m·s ˉ1, the variation of produc-tions of gasoline and gases with the inlet gas velocity are not noticeable. It is expected that the clusterbreakup may occur when the inlet gas velocity reaches some point. However the cluster breakage is not con-sidered in present model.3.4 Effect of operating temperatureIncreasing the operating temperature has a posi-tive effect on cracking reactions. Fig. 11 shows the reaction rates of the cluster and an isolated particle as a function of inlet gas temperature. The reaction rates from VGO to gasoline, gas and coke of the isolated particle are higher than those of the cluster. While the reaction rates of gasoline to gas and coke of the cluster are larger than those of the isolated particle. Since the reaction rate from gasoline to gas is higher than that from gasoline to coke, more gasoline will be con-verted to gas.The yield profiles of gas and gasoline are shown in Fig. 12 as a function of inlet gas temperature. The reaction rates were increased as the temperature re-sults in more gas and gasoline produced. For the iso-lated particle, the gas and gasoline yields incrementswith the increase of temperature are expected. The gasFigure 7 Reaction rates as a function of cluster porosity (u g ˙1 m·s ˉ1, T ˙991 K, mass fraction of VGO ˙0.27, mass fraction of steam ˙0.73, number of isolated particle ˙65)Figure 9 Profile of mass fluxes of gas and gasoline as a function of inlet velocity (mass fraction of VGO ˙0.27, mass fraction of steam ˙0.73, İg ˙0.8; T ˙991 K) Ƶ average value of cluster; ƻisolated particleFigure 10 Profile of mass fluxes of gas and gasoline as a function of inlet temperature of mixture gases [Both numbers of isolated particles and particles in cluster are 65, mass frac-tion of VGO ˙0.27 (or concentration of VGO ˙0.2 mol·m ˉ3), mass fraction of steam ˙0.73, İg ˙0.8, T ˙991 K] isolated particle: Ƶ gas; Ʒ gasoline cluster: ƶ gas; Ƹ gasolineChin. J. Chem. Eng., Vol. 16, No. 5, October 2008676production increases much more rapidly than thegasoline production as the temperature increases. However because of the produced gasoline is partially converted to gas by reaction route D, an optimal oper-ating temperature may need to be determined for products selectivity.3.5 Effect of inlet mass fraction of VGOThe cracking reaction rate for route A is propor-tional to the square of the concentration of VGO, while for other reactions the rates are proportional to the concentration of VGO. Hence, the reaction rate increases with the increase of inlet mass fraction of VGO. Fig. 13 shows the reaction rates of the cluster and an isolated particle as a function of inlet mass fraction of VGO. The reaction rates from VGO to gaso-line, gas and coke of the isolated particle are higher than those of the cluster, while the reaction rates from gasoline to gas and coke of the cluster are higher than those of the isolated particle. This indicates that the particle clustering will reduce VGO cracking.Figure 14 shows the distributions of gas and gasoline as a function of inlet mass fraction of VGO.It is noticed that more gasoline produced and less gas production from the isolated particle than those from the individual particle in the cluster. Simulated results indicated that particle clustering will reduce the produc-tion of gasoline and increase the yield of gas in the riser. 3.6 Effect of coke activityIn the present model, the effect of coke deposited on the catalyst is considered by coke concentration C c in Eq. (11). With the increase of C c , the value of activ-ity factor Mdecreases. The cracking reactions of catalyst particle clusters were simulated with different values of the coke concentrations to understand the effect of coke formation on cracking reactions of the clusters. Fig. 15 shows the reaction rates of the cluster and an isolated particle as a function of coke concen-trations. With the formation of coke, the reaction rates of the cluster and the isolated particle are decreased.The predicted mass fluxes of gasoline and gas of the cluster and an isolated particle are shown in Fig. 16 as a function of coke concentrations. It can be seen that when a value 0 for C c was selected, the mass fluxes of gas and gasoline are highest due to the factthat the activity factor equals to unity without cokeFigure 11 Profile of mass fluxes of gas and gasoline as a function of inlet velocity (mass fraction of VGO ˙0.27, mass fraction of steam ˙0.73, İg ˙0.8, T ˙991 K) Ƶ average value of cluster; ƻisolated particle Figure 12 Profile of mass fluxes of gas and gasoline as afunction of inlet temperature of mixture gases [Both num-bers of isolated particles and particles in cluster are 65, u g ˙1 m·s ˉ1, İg ˙0.8, mass fraction of VGO ˙0.27 (or concentra-tion of VGO ˙0.2 mol·m ˉ3), mass fraction of steam ˙0.73] isolated particle: Ƶ gas; Ʒ gasoline cluster: ƶ gas; ƸgasolineFigure 13 Profile of mass fluxes of gas and gasoline as afunction on inlet molar concentration of VGO (u g ˙1.0 m·s ˉ1,İg ˙0.8, T ˙991 K) Ƶ average value of cluster; ƻisolated particleFigure 14 Effect of coke contents on mass fluxes of gasand gasoline (Both numbers of isolated particles and particles in cluster are 65, u g ˙1 m·s ˉ1, İg ˙0.8, T ˙991 K) isolated particle: Ƶ gas; Ʒ gasoline cluster: ƶ gas; Ƹ gasoline。
A Discrete Binary Version of the Particle Swarm Optimization Algorithm
where cp is a random positive number generated for each id, whose upper limit is a parameter of the system, and: Xid = Xid + “id The particle swarm algorithm has been found to be robust in solving problems featuring nonlinearity and nondifferentiability, multiple optima, and high dimensionality through adaptation which is derived from social-psychologicaltheory (Kennedy, 1997).
0-78034053-1/!l7/$10.OO
1997 IEEE
4104
with zero bits flipped does not move, while it moves the “farthest” by reversing all of its binary coordinates. This does not answer the question, however, of what corresponds to vi+ in a binary function, that is, what is the velocity or rate of change of a single bit or coordinate. The solution to this dilemma is to define trajectories, velocities, etc., in terms of changes of probabilities that a bit will be in one state or the other. Thus, a particle moves in a state space restricted to zero and one on each dimension, where eaclh Vid represents the probability of bit Xid taking the value 1. In other words, if Vid = 0.20, then there is a twenty percent chance that Xid will be a one, and an eighty percent chance it will be a zero. If the previous best positions have had a zero in that bit, then @id - Xid) can be reasonably calculated as -1, 0, or +1, and used to weight the change in probability Vid at the next step. In sum, the particle swarm formula: “id = Vid + (P(Pid- Xid) + HPgd - Xid) remains unchanged, except that now p d and Xid are integers in (0, l } and Vid, since it is a probability, must be constrained to the interval1 [O.O, 1.01. A logistic transformation s(Vid) can be used to accomplish this last modification. The resulting change in position then is defined by the following rule: $(rand() < s(vid)) then Xid = 1; else Xid = 0 where the function S(v) is a sigmoid limiting transformation and rand() is a quasirandom number selected from a uniform distribution in [O.O, 1.01. The continuous-valued particle swarm algorithm also limited v d by a value V,, which ‘was a parameter of the system. In the discrete vlersion V,, is retained, that is, Ivid < V-, but as can be seen, this simply limits the ultimate probability that bit Xid will take on a zero or one value. For instance, if V d 6 . 0 , then probabilities will be limited to s(Vid), between 0.9975 and 0.0025.. The result of this is that new vectors will still be tried, even after each bit has attained its best position. Specifying a e.g., 10.0, makes new vectors less higher V,,, in the likely. Thus part of the function of V,, discrete particle swarm is to set a limit to further exploration after the population has converged; in a sense, it could be said to control the ultimate mutation rate or temperature of the bit vector. Note also that, while high V,, in the continuous-valued version increases the range explorad by a particle, the opposite occurs in the binary version; smaller V,, allows a higher mutation rate.
高能量密度物理国际会议邀请报告-许爱国
D /2
mc 2 exp 2 KT
where c v u
f f 1 v ( f f eq ) t r
Lattice Boltzmann Method
【8】
…
eq
f
n
2 KT
D /2
mc 2 exp 2 KT
Claude-Louis Navier
Leonhard Euler
George Gabriel Stokes
【5】
The concerned high Mach number compressible flows
are generally related to shock waves.
R M
S
l Kn L
Numerical solvers of NS or Euler equations, little applications.
【13】
2006, Shan et al, a theoretical framework for representing hydrodynamic system through a systematic discretization of the Boltzman Eq. by means of Hermite tensor expansion of the Maxwellian distribution function.
Hale Waihona Puke The BGK (Bhatnagar-Gross-Krook)
is one of the
f
Max Krook (died 4 August 1985) was an American m athematician and astrophysicst
废弃塑料实现木塑复合材料的制备及其性能探究
第43卷 第11期 包 装 工 程2022年6月PACKAGING ENGINEERING ·24·收稿日期：2021–08–24基金项目：山东省自然科学基金（ZR2019BC108，ZR2019BC021）；大学生创新创业项目（S201910431045）；国家自然科学基金（31901273）；生物基材料与绿色造纸国家重点实验室项目（ZZ20190204，ZZ20190201） 作者简介：张志礼（1988—），男，博士，齐鲁工业大学讲师，主要研究方向为生物基材料及应用。
废弃塑料实现木塑复合材料的制备及其性能探究张志礼，王新婷，李凤凤，褚夫强（齐鲁工业大学 生物基材料与绿色造纸国家重点实验室，济南 250353）摘要：目的 基于桉木粉（Wood, W ）和废弃聚乙烯塑料瓶（PE ）实现木粉/PE 复合材料（WPE ）的制备，并对其性能进行探究。
方法 借助塑料注射成型机实现WPE 的制备，通过外观分析、拉伸试验、TGA 谱图、HalpinTsai 模型分析和生物降解性研究WPE 的性能。
结果 桉木粉在PE 基体中能够实现良好分散，且当木粉质量分数为3%时，WPE 的弹性模量和断裂伸长率分别为1.69 GPa 和153%，此时WPE 质量损失5%时的分解温度（t 5%）和最大分解速率温度（t max ）分别提高了5.13 ℃和2.73 ℃。
土埋320 d 后，WPE 的最大质量损失率为20.25%，远高于PE 的质量损失率（2.35%），充分表明增加木粉可以显著提高木塑复合材料的生物降解性。
结论 借助注射成型法可成功实现WPE 的制备，且木粉较佳质量分数为3%。
关键词：废弃塑料；木粉；木塑复合材料；生物降解性中图分类号：TB332 文献标识码：A 文章编号：1001-3563(2022)11-0024-07 DOI ：10.19554/ki.1001-3563.2022.11.004Preparation and Properties of Wood-plastic Composites Based on Waste PlasticsZHANG Zhi-li , WANG Xin-ting , LI Feng-feng , CHU Fu-qiang(State Key Laboratory of Biobased Material and Green Papermaking, Qilu University of Technology, Jinan 250353, China) ABSTRACT: Based on eucalyptus powder (Wood, W) and waste polyethylene plastic bottles (PE), the preparation of wood powder/PE composite (WPE) and its performance are explored. The WPE was prepared by plastic injection molding machine, and the performance of WPE was studied through appearance analysis, tensile test, TGA spectrum, HalpinTsai model analysis and biodegradability. Eucalyptus powder can be well dispersed in the PE matrix. When the wood power mass is 3wt.%, the elasticity modulus and elongation at break of WPE are 1.69 GPa and 153% respectively. At this time, the temperature at which WPE decomposes by 5% (t 5%) and the maximum decomposition rate temperature (t max ) in-crease by 5.13 ℃ and 2.73 ℃ respectively. After 320 days of soil burial, the maximum mass loss rate of WPE is 20.25%, which is much higher than the mass loss rate of PE of 2.35%, which fully shows that increasing wood powder can signif-icantly improve the biodegradability of wood-plastic composites. The preparation of WPE can be successfully achieved by the injection molding method, and the preferred amount of wood powder is 3wt.%. KEY WORDS: waste plastic; wood powder; wood-plastic composite; biodegradability木塑复合材料作为一种新型的绿色环保的复合材料[1-2]，凭借其力学性能好、耐腐蚀、防水、防虫蛀、便于加工等优点引起了广泛关注[3-4]。
有机溶剂挥发量之估算方法
有机溶剂挥发量之估算方法赵焕平中原大学生物环境工程学系32023桃园县中坜市中北路200号Tel: 03-2654914，Fax: 03-2654949，E-mail有机溶剂常具有高挥发性与毒性，当它们挥发至大气中，往往会对工厂操作人员或附近居民健康造成威胁，许多工厂考虑以排放系数计算挥发性有机物(VOCs)之挥发量，虽然使用排放系数可以概略的估算出工厂整体挥发性有机物的排放量，但若要考虑有机溶剂挥发对人体健康可能产生之影响，则需要针对不同有机溶剂的个别挥发量进行估算。
可惜的是，目前所有对于有机溶剂挥发所发展出的方法均为经验方程式，虽然可以在特定条件下估算有机溶剂的挥发量，但往往因为环境条件改变或工作地点变更而造成估算误差，本文的主要目的在提供一个估算有机溶剂挥发量的方法，希望藉由此方法能够估算工厂内有机溶剂之挥发量，以减少有机溶剂周遭工作劳工可能产生之危害。
一、影响有机溶剂挥发之参数有机溶剂的挥发主要受到其本身性质的影响，其中最主要的影响因子为有机物本身的饱和蒸气压，常见有机化合物的蒸气压可以由一般物理化学的工具书查到[1]，若是针对同一类型的有机物，如烷类、醇类、烯类等，饱和蒸气压将会随有机物分子量增加而减少。
若是针对不同类型的有机物如烷类与醇类，尽管分子量接近，但醇类分子间有氢键，因此戊醇的蒸气压将远低于戊烷，除了有机溶剂本身的性质外，环境因子也会影响有机溶剂的挥发量，其中最明显者为温度与风速[2,3]，温度改变有机物的饱和蒸气压也会改变，通常温度增加有机物的饱和蒸气压会增加，特别是高蒸气压的化合物，温度增加时，蒸气压增加的趋势更为明显。
另外液面上方风速也是一个非常重要的影响因子，有风与无风的条件下有机物的挥发量相差常超过10倍，因此当有风通过液面上方，将使有机溶剂的挥发量明显增加[4]。
除上述影响因子外，尚有一些因子可能会影响挥发量，但却常常被忽略不计，如液体搅拌就是一个明显的例子，一般观念认为纯的有机溶剂进行搅拌，并不会增加它们的挥发量，但实际上，搅拌可使有机溶剂表面产生扰动，仍然会使有机溶剂的挥发量增加，只是此挥发增加有限，因此除非是非常迅速搅拌，否则可以忽略不计。
1-s2.0-S0360544211003513-main
Optimal sizing of a solar thermal building installation using particle swarm optimizationRaffaele Bornatico a ,*,Michael Pfeiffer a ,b ,Andreas Witzig b ,Lino Guzzella aa Institute for Dynamic Systems and Control ETH Zurich,Sonneggstrasse 3,CH 8092Zurich,Switzerland bVela Solaris AG,CH 8400Winterthur,Switzerlanda r t i c l e i n f oArticle history:Received 8October 2010Received in revised form 6May 2011Accepted 15May 2011Available online 30June 2011Keywords:Particle swarm optimization Multi-objective optimization PolysunSolar combisystem Solar energy Smart buildinga b s t r a c tIn recent years the domestic energy management has become a non-trivial task as the number of energy sources and system components involved have increased,and all components have to operate coordi-nately in order to maximize global ef ficiency measures.In this paper a methodology is presented for finding the optimal size of the main components for a solar thermal system where particular attention is given to the optimization framework.The use of the PSO (particle swarm optimization)algorithm is proposed and the results obtained are compared with a GA (genetic algorithm)solution.Further,the relative in fluence of certain system parameters on the optimal con figuration is investigated by means of a sensitivity analysis where the size of the collector is shown to have the greatest in fluence on all main output quantities while the size of the auxiliary power unit presents a relatively small in fluence on the solution.Finally,it is demonstrated that the accurate sizing of the energy components is necessary to minimize the energy consumption and cost of installation,while maximizing the solar fraction.The proposed methodology is shown to successfully solve the problemÓ2011Elsevier Ltd.All rights reserved.1.IntroductionSolar energy for SH (space heating)and the production of DHW (domestic hot water)has become an important factor in reducing the global CO 2emissions.Within the previous and the last United Nations climate change conference it has been analyzed what needs to be done to limit the long-term concentration of green-house gases in the atmosphere to 450ppm of CO 2equivalent,in line with a 2 C increase in global temperature by 2100.Solar energy is considered as one of the most promising candidates to tackle our dependency on,and use of fossil fuels,and thus for reducing the related emissions.The IEA (International Energy Agency)with its “Task 26”on solar combisystem emphasizes that if the direct use of solar energy is to make a signi ficant contri-bution to the heat supply,it is necessary that solar-heating tech-nologies must be developed and widely applied over and beyond the sole field of DHW preparation [1,2].Accordingly,the focus of this research has been set on a solar combisystem installation that simultaneously ful fills DHW and space heating needs.Previous publications on solar combisystems have presented analyses ofthe effects of geographic location [3]and the use of different DHW load pro files [4]on solar fraction and energy consumption of the system.This paper considers a mid-sized single-family house located in Zurich,Switzerland,and presents a methodology for finding the optimal component sizes for a solar combisystem.All simulations are carried out with Vela Solaris Polysun Ò,which is a well-established software tool in the field of planning and optimiza-tion of building energy systems.The simulation kernel is based on a plug-flow simulation of the thermal system [5],and it uses statistical meteorological data as an input [6,7].The optimization routine was implemented in MATLAB Ò.The interface from MATLAB to the simulation kernel is called Polysun Inside.Polysun Inside is a Polysun plug-in that allows all simulation functionality to be controlled from the MATLAB environment in an automatic-iterative evaluation routine.In the speci fic case,a par-allelization over 10CPUs has been performed which considerably decreased the optimization time.This paper is structured as follows:In Section 2the modeling assumptions and the simulation setup are presented.Section 3describes the optimization framework and all parameters neces-sary for the optimization routine to succeed.Section 4describes the main features of the PSO algorithm and introduces the subsequent results sections.*Corresponding author.Tel.:þ41446322453;fax:þ41446321139.E-mail address:rbornatico@ethz.ch (R.Bornatico).Contents lists available at ScienceDirectEnergyjournal h omepage:w/locate/energy0360-5442/$e see front matter Ó2011Elsevier Ltd.All rights reserved.doi:10.1016/j.energy.2011.05.026Energy 41(2012)31e 372.Solar combisystem model in PolysunThe simulation software Polysun offers a broad range of func-tionalities required for the analysis and design of domestic energy systems.In the Polysun catalogs,a variety of components are available with all characteristic data and efficiency maps necessary for the hydronic and thermal analysis of such systems.Analysis and design of photovoltaic systems,solar cooling,and combinations of solar thermal and heat pump systems are also possible.Polysun calculates all relevant system parameters related to the production of heat and electricity.It also comprises the calculations for system amortization and the data required for subsidy appli-cations.The variable step solver adapts the simulation step size to capture the transient effects down to a minimum of4min.Itfollows that sufficient accuracy of the system’s simulated charac-teristics and in particular of its dominant transients can be assumed.The user friendly interface of Polysun allows for the easy parameterization of the system,while the software itself is capable to output and store practically all relevant physical quantity in convenient data formats.Furthermore a mean simulation time of 1min for a one-year simulation makes the software Polysun an ideal platform to perform the mentioned analyses[7].The setup of the solar combisystem used in this research is depicted in Fig.1.Leftmost is the APU(auxiliary power unit)and rightmost is the solar collector.Both components are connected to the storage tank in the center of the picture.The loads are char-acterized by a DHW and a space heating demand.The DHW profile used divides200l/d of water at45 C into the periodic daily demand depicted in Fig.2.The effects of using a realistic profile as opposed to a represen-tative periodic profile are discussed in detail in Ref.[4],where it is shown that the heat demand difference between the two approaches is0.1%and the fractional energy saving difference is just around1.5%when using aflow regulation device.The greatest influence is shown to be caused by the reduced demand during the summer holidays in the realistic profile.Considering that with Polysun absences can be included in the simulations,it is reason-able to assume that the profile used hereafter approximates suffi-ciently well a real DHW load.Energy for space heating is less straightforward to determine since it depends on current ambient temperature and building insulation.The heating setpoint temperature is20 C during the day and18 C at night.Other simulation relevant parameters of the building include the U-value of0.45;the heated area of150m2; the air change rate of0.3hÀ1;the air infiltration of0.6hÀ1;the internal heat gains of2W mÀ2;the heat gains from equipment of 240W;the heat capacity of500kJ KÀ1mÀ2and the solar heat gain coefficient of0.8.The collector is oriented toward South(0 )with an inclination angle of45 .The resulting simulated heat demand is shown in Fig.3and introduces the importance of the availability of comprehensive meteorological data.2.1.Statistical meteorological dataTo a large extent,the accuracy of solar system simulations for a given location depends on the availability of realistic data of solar irradiation,humidity,etc.In Polysun,meteorological data are provided by the Meteonorm database[8]containing data based on measurements from8055weather stations worldwide.For any given location,the data of the closest weather stations are interpolated.The generation of yearly series of weather data utilizes stochastic models, where stored monthly mean values and Markov Transition Matrices are used to generate hourly weather data[9,10].The resulting hourly data have the same statistical properties as the measured data(i.e. average value,variance,autocorrelation),and thus represent an accurate approximation.In Fig.4,the outdoor temperature and global irradiance time series are shown over a one-yearperiod.Fig.1.A solar combisystem designed usingPolysun.Fig.2.Daily domestic hot water heat demand.R.Bornatico et al./Energy41(2012)31e37323.Optimization frameworkThe main components of the system as presented in the previous section are the collector,the tank,and the APU.In order for these components to be optimally dimensioned,the corre-sponding sizes have to populate the input parameter set that is fed to the simulation kernel.In Table1the input set is listed together with the corresponding minimal and maximal values used throughout this work.The extreme values for the output set shown in Table2do not represent a given constraint.Rather,these values have been obtained from simulations throughout this work.They are to show that the investigated range for these values is significant.The goal of the optimization routine consists infinding the parameters w opt that minimize a general function f:R n/R that is formulated in the canonical form as follows:min w fðwÞsubject to gðwÞ<0;hðwÞ¼0(1)The functions g and h include all constraints to the problem defining the set Q of the acceptable w˛Q.These constraints include the boundary values of w and two physical constraints on the system.First,for a given w to be acceptable,the heat requirements for DHW and space heating must be satisfied.Secondly,the temperature of thefluidflowing in the collector must not exceed the100 C threshold for longer than1%of the simulation time.This last condition limits the stagnation time to reasonable values, thereby ensuring the safe operation and a long life of the collector.In the presented problem a numeric optimization is necessary because an algebraic solution to the problem does not exist. Furthermore it is clear that the topographical properties of the function f play a crucial role in how the optimization algorithm converges from starting values w0toward w opt,and also that the choice of the right cost function is determinant for the successful convergence of the algorithm to the point of interest.Accordingly, all relevant quantities should appear in the cost function.In the case of the optimization of the combisystem,the objectives of the authors are to maximize the solar fraction,to minimize the total energy use,and to minimize the additional cost of the installation.fðwÞ:¼8<:y r y¼Xjw j$f jðwÞj max9=;(2)Therefore,the rather intuitive linear form(2)is chosen where the solar fraction f1ðwÞ¼SFnðwÞ,the total energy use of the system f2ðwÞ¼EtotðwÞand the cost of the installation f3ðwÞ¼CostðwÞdepend on the sizes of the components w.Since a scalar value of f is required by the minimization algo-rithm,the weighted sum,with weights w i of the normalized f i, yields a dimensionless,normalized cost function fðwÞ.Note that in order to maximize the solar fraction by a minimization of f,w1must be negative.An important result from a user’s point of view is the pricing of the components.This term is based on a linear pricing assumption for each component and on current oil prices.This choice is shown in Ref.[11]to produce meaningful results.The relevant values used for the installation pricing are listed in Table3.4.The particle swarm optimization algorithmThe analogy of a population of individuals,called particles(e.g.a swarm of bees)having the common goal offinding an optimal position(e.g.the bestflower in afield)is useful to picture how the algorithm converges toward a solution.This socially inspired optimization technique wasfirst introduced by Kennedy and Eberhart[12]as a promising tool that does not requires the computation ofderivatives.Fig.3.Yearly heat requirements for space heating in a mid-sized single-familyhouse.Fig.4.Meteorological data for Zurich,Switzerland.Solid:outdoor temperature,dotted:global irradiance.R.Bornatico et al./Energy41(2012)31e3733Each particle i updates its position toward better fitness areas in the problem domain landscape based on a self cognitive term expressed by the best position (P i )found so far by the particle,and the global best position encountered so far by the entire swarm (G ).Starting with random initial velocities v 0i and random initial posi-tions p 0i inside a closed set,the basic PSO algorithm computes the velocity and position of the i -th particle,i ¼1;2;.;n for every iteration k ¼0;1;2;.k max in the following way:v k þ1i¼v k i þg 1iP i Àp k iþg 2iG Àp k i(3)p k þ1i¼p k i þv k þ1i (4)The random numbers g 1;2˛½0;1 in fluence the magnitude of thetwo vectors ðP i Àp k i Þand ðG Àp k i Þ,thus in fluencing indirectly the magnitude and direction of v k þ1i.Fig.5depicts how the update from k to k þ1is performed for a simpli fied two dimensional problem with no randomness g 1,2¼1.Note that every dimension represents a parameter to be optimized.The common PSO algorithm includes three additional parame-ters,namely an inertia function f k and two acceleration constants a 1,2.The inertia function used hereafter is linearly decreasing with respect to iterations,thus reducing the in fluence of past velocities,thereby enabling the algorithm to adapt to small regions as the optimization converges.This adaptation capability is of key rele-vance since it makes the use of hybrid methods unnecessary.The resulting velocity equation,together with the position update,represents the core of the PSO algorithm [13,14].v k þ1i¼f k v k i þa 1h g 1i P i Àp k i i þa 2h g 2i G Àp k i i(5)p k þ1i¼p k i þv k þ1i (6)Further extensions of the algorithm can include natural selection considerations such as a varying population size where bad parti-cles are killed and new particles are generated in the vicinity of the global best point.Dynamic adjustment of swarm parameters have been also proposed as a way to avoid local optima [15e 17].The work of Clerc and Kennedy [18]describes the importance of including constraints on the velocity vector.Their work shows that if no constraints are set the algorithm can become unstable.Note that all these extensions increase the computational complexity of the algorithm and should therefore be avoided,if not necessary.In the problem presented here a fixed population of 10particles has been deemed to be suf ficiently effective.Further,a linearly decreasing inertia function already accounts for stability,as the introduction of velocity limits can be shown to be a loose constraint.5.Optimization resultsThe results reported in this section have been based on simu-lation results of the software Polysun when simulating the model presented in Section 2.The converged results of the minimization of Eq.(2)by means of the PSO algorithm are presented in Section 5.1,while results obtained using a genetic algorithm follow in Section 5.2.A further investigation on the solution is presented in the form of a sensitivity analysis,which follows in the subsequent Section 6.5.1.Particle swarm optimization:resultsThe minimization of the fitness function by means of the PSO algorithm is shown in Fig.6.The fitness function is plotted against particle generations where each generation consists of 10particles.Note that points which do not satisfy the optimization constraints have been omitted.The PSO algorithm converges toward a minimal fitness value of f opt ;PSO ¼1:278and the interpretation of this value,with respect to its physical terms,is presented in Table 4.The initial population was randomly generated inside the boundary setTable 2Output parameter set.ItemSymbol Min Max Unit Solar fraction f 1 2.8%42.7%e Energy use f 21298845407kW h Costf 31298846226VTable 3Relevant factors for the installation pricing.Item Value Unit Collector 100V Collector 360V /m 2Storage tank 4.22V /l APU50V /kW Heating oil 10.5kW h/l Heating oil 0.85V /l Maintenance100V/yearFig.5.Simpli fied PSO iteration in twodimensions.Fig.6.Fitness function minimization with PSO.Table 1Input parameter set.ItemSymbol Min MaxUnit Collector area w 1140m 2Tank volume w 21003000l APU sizew 3550kWR.Bornatico et al./Energy 41(2012)31e 3734Q .It follows that the starting values as reported in Table 4,areaveraged values of the feasible points of the initial population.It is important to judge the convergence of the algorithm by considering the spread of the particles.Fig.7illustrates the convergence of all particles,for all parameters considered in this study.From Figs.6and 7it can be seen that after 50generations the algorithm has converged to the vicinity of the optimal values which are found to be w opt ;PSO ¼f 14:5m 2;498:98l ;8:5kW g .5.2.Genetic algorithm optimization:resultsOn the same problem a standard GA has been applied where,if applicable,the same optimization parameters as in the PSO were used,speci fically,a population of 10particles and the same parameter set boundaries.In this case also,the fitness function decreases toward a minimal value of f opt ;GA ¼1:278,a value that is consistent with the result of the PSO optimization.The compar-ison of all other optimal fitness function terms between PSO (Table 4)and GA (Table 5)shows that the two solutions differ by a minimal amount only.Also for the GA,the convergence of the fitness function and the system parameters is shown in Figs.8and 9.Fig.8shows that the convergence of all parameters over generation number is compa-rable to those obtained with the PSO optimization as the algorithm converges toward an optimal value of w opt ;GA ¼f 15:1m 2;514l ;8:6kW g .A slight difference in the optimal parameters obtained for the collector area and the tank volume can be noticed.However,this does not in fluence signi ficantly the fitness function values,thus proving that both algorithms converge to a similar solution.Giventhat the search space is rather big and that the starting individuals are randomly drawn inside this space,the fact that the optimal values found with GA and PSO are very similar supports the consistency of the results and the solution found in this section.6.Parameter sensitivity analysisThe results in Section 5are speci fic for the considered system,i.e.a daily DHW consumption of 200l/day at 45 C and an SH setpoint temperature of 20and 18 C during day and night respectively.First the sensitivity of the results with respect to a variation of these parameters is explored and secondly,the same analysis is performed over the optimization parameters w 1;2;3.All results in this section are produced using the optimal PSO values w opt ;PSO iand f opt ;PSOj as nominal values when necessary.6.1.Operational parameter variationThus far,constant operational parameters have been used.Among them,the amount of DHW consumed (p 1)and its temper-ature (p 2)together with the heating setpoint temperature of the building during day (p 3)and night (p 4)are investigated here.Initially the DHW volume p 1is varied between {50,100,200,300,400}l/day and the simulation outcomes are linearly fitted to produce the results shown in the first column of Table 6.For instance the ratio ðv SFn Þ=ðv p 1Þis 0.01[%/l]indicating a solar fraction increment of 1%for every additional 100l/day DHW consumed on top of the nominal 200l/day.Similarly,the DHW temperature (p 2)isvaried between {35,40,45,50,55} C,p 3˛{18,19,20,21,22}C andp 4˛{16,17,18,19,20}C to produce the other values in Table 6.The sensitivity of the results relative to the corresponding nominal values is also a useful mean of comparison.The relative sensitivity computed using Eq.(7)is graphically shown in Fig.10where p nom ¼{200l/day,45 C,20 C,18 C}.The relative varia-tion of the daily heating setpoint temperature clearly dominates all other effects and indicates that this parameter is of crucial impor-tance in the characterization of the system.From this figure it can be read that a 1%increase in p 3causes an additional energy consumption of around 2%.S ij ¼v f j=f opt ;PSO j p i p nomi(7)Table 4PSO optimization of fitness function terms.Term Starting value Optimized value Change f 1.435 1.278À10.1%SFn 18.55%21.8%þ17.5%Etot 16428kW h 15806kW h À3.8%Cost13210V8963VÀ32.1%Fig.7.PSO convergence of system parameters towards 14.5m 2(a),499l (b),and 8.5kW (c).Table 5GA optimization of fitness function terms.Term Starting value Optimized value Change f 1.335 1.278À4.2%SFn 20.6%22.15%þ7.5%Etot 16109kW h 15750kW h À2.2%Cost10952V9230VÀ15.7%Fig.8.Fitness function minimization with GA.R.Bornatico et al./Energy 41(2012)31e 37356.2.Optimization parameter variationIn contrast to the previous section where operational parame-ters are varied for a given installation,this section deals with varying size of collector,tank volume and APU power.Each one of these parameters is varied by {À10%,À5%,þ5%,þ10%}of the respective nominal values,which correspond to the optimal PSO results w opt ;PSO ¼f 14:5m 2;498:98l ;8:53kW g .The same methodology as in the previous section is applied here and the results are presented in absolute values in Table 7and in Fig.11according to Eq.(8).S ij ¼v f j=f opt ;PSOj v ðw i Þ=w i(8)The relative sensitivity of the solar fraction S i 1is depicted in Fig.11by the bars (1)where it can be seen that an increase of the collector or tank size has a positive effect on the solar fraction.This effect is dominated in relative terms,by the solar collector with S 11¼0.35.Further,bar (2)of w 3shows that an increase in APU size in the vicinity of the optimal solution causes a greater energy consumption of the system due to a shifted operating point in the APU consump-tion and ef ficiency map,whereas an increase of collector size or tank volume reduce the overall energy consumption of the system.The bars labeled (3)in Fig.11show that an increase in collector size has the greatest relative in fluence on the cost of the installa-tion.The tank volume has a smaller in fluence,while the in fluence of the APU size is relatively low.7.Conclusions and discussionIn this paper the optimal sizing of a solar thermal system has been presented.The analysis is performed on a solar combisystem for a mid-sized single-family house in Zurich,Switzerland.While the optimization framework is in principle independent on the optimization algorithm,a detailed analysis has been carried out on the performance of the Particle Swarm Optimization algorithm when it is applied for solving this problem.The results are comparable to those obtained with the more common Genetic Algorithm.When the implementation efforts and the computa-tional power demand are considered as well,the PSO is a slightly better choice for solving the presented problem.A collector size of 14.5m 2together with a tank volume of 498.98l and an APU nominal power of 8.5kW,are the optimal sizes for the main system ’s components,which lead to a solar fraction of 21.8%,a total energy use of 15806kW h and a cost of the installation of 8963V .The parameter sensitivity analysis shows that the size of the collector has the greatest in fluence on the solar fraction,energy use and the installation cost,while the tank volume in fluence is signi ficant on the solar fraction and cost of installation.The size of the APU has a relatively small effect on the both energy use and installation cost.The variation of selected operationalparametersFig.9.GA convergence of system parameters towards 15.1m 2(a),514l (b),and 8.6kW (c).Table 6Absolute sensitivity of operational parameters.p 1[l/day]p 2[C]p 3[C]p 4[C]SFn [%]0.010.02À1.14À0.24Etot [kW h]9.670.61638.591.3Cost [V ]0.735.16125.667.69Fig.10.Relative sensitivity of the optimal solar fraction (1),total energy consumption (2)and cost of installation (3)related to a variation of DHW volume p 1,DHW temperature p 2,SH day setpoint temperature p 3and SH night setpoint temperature p 4.Table 7Absolute sensitivity of optimization parameters.w 1½m 2w 2½l w 3½kW SFn [%]0.530.01À0.3Etot [kW h]À94.4À0.78119.30Cost [V ]352.94.260.6Fig.11.Relative sensitivity of the optimal solar fraction (1),total energy consumption (2)and cost of installation (3)related to a variation of collector size w 1,tank volume w 2,and rated power of the auxiliary power unit w 3.R.Bornatico et al./Energy 41(2012)31e 3736has been investigated as well,and shows that the sensitivity of the daily heating setpoint temperature of the building is by far the dominant parameter when compared to all parameters considered in this study.Future work includes an extension and further investigation of the parameter set.Furthermore,the proposed methodology is to be applied over other solar installations and/or more complex models including,for example a detailed life cycle analysis of the system. AcknowledgmentsThis work has been partly funded by the Swiss Innovation Promotion Agency CTI,whose support is gratefully acknowledged. References[1]Sawin E,Jones A,Sterman J.Final Copenhagen accord press release19december09-expanded version;2009.[2]Suter JM,Letz T,Weiss W,Inbnit J.Solar combisystems in Austria,Denmark,Finland,France,Germany,Sweden,Switzerland,the Netherlands and the USA overview2000;2000.[3]Lund P.Sizing and applicability considerations of solar combisystems.SolarEnergy2005;78(1):59e71.[4]Jordan U,Vajen K.Influence of the dhw load profile on the fractional energysavings:a case study of a solar combi-system with trnsys simulations.Solar Energy2001;69(Suppl.6):197e208.[5]Klein S,Beckmann B,Duffie J.Trnsys,a transient system simulation program,users manual,version16;2006.[6]Duffie J,Beckman W.Solar engineering of thermal processes.3rd ed.WileyInterscience;2006.[7]Witzig A,Foradini F,Probst MCM,Roecker C.Simulation tool for architects.In:Proceedings of the2009CISBAT international conference;2009.[8]Meteotest.Meteonorm database.Bern,Switzerland:Meteotest;2000.[9]Meteotest.Meteonorm handbook,part2:theory.Bern,Switzerland:Meteotest;2000.[10]Aguiar RJ,Collares-Pereira M,Conde JP.Simple procedure for generatingsequences of daily radiation values using a library of Markov transition matrices.Solar Energy1988;40(3):269e79.[11]Fraisse G,Bai Y,Pierrs NL,Letz parative study of various optimizationcriteria for sdhws and a suggestion for a new global evaluation.Solar Energy 2009;83(2):232e45.[12]Kennedy J,Eberhart R.Particle swarm optimization.In:Proceedings of the1995IEEE International Conference on Neural Networks,vol.4;1995.p.1942e1948.[13]Birge B.Psot e a particle swarm optimization toolbox for use with matlab.In:Proceedings of the2003IEEE Swarm Intelligence Symposium SIS03;2003.p.182e186.[14]Bratton D,Kennedy J.Defining a standard for particle swarm optimization;2007.120e127.[15]Dimopoulos GG,Frangopoulos CA.Optimization of energy systems based onevolutionary and social metaphors.Energy2008;33(2):171e179.19th Inter-national Conference on Efficiency,Cost,Optimization,Simulation and Envi-ronmental Impact of Energy Systems e ECOS2006.[16]Shi Y,Eberhart R.A modified particle swarm optimizer.In:Proceedings ofthe1998IEEE World Congress on Computational Intelligence;1998.p.69e73.[17]Fourie P,Groenwold A.The particle swarm optimization algorithm in size andshape optimization.Structural and Multidisciplinary Optimization2002;23: 259e67.[18]Clerc M,Kennedy J.The particle swarm e explosion,stability,and conver-gence in a multidimensional complex space.IEEE Transactions on Evolu-tionary Computation2002;6(1):58e73.R.Bornatico et al./Energy41(2012)31e3737。
- 1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。
Available online at Journal of Power Sources174(2007)949–953Short communicationEffect of particle size on LiMnPO4cathodesThierry Drezen a,Nam-Hee Kwon a,Paul Bowen b,Ivo Teerlinck c,Motoshi Isono d,Ivan Exnar a,∗a High Power Lithium(HPL),Scientific Park B(PSE-B),Swiss Federal Institute of Technology,Lausanne,Switzerlandb Laboratoire de Technique des Poudres(LTP),Institut des Mat´e riaux,Facult´e des Sciences et Techniques de l’ing´e nieur,Ecole,Polytechnique F´e d´e rale de Lausanne,Lausanne,Switzerlandc Toyota Motor Europe NV/SA,Hoge Wei33,B-1930Zaventem,Belgiumd Material Engineering Division,Toyota Motor Corporation,Toyota-cho,Toyota-shi,Aichi471-8571,JapanAvailable online30June2007AbstractLiMnPO4was synthesized using a sol–gel method and tested as a cathode material for lithium ion batteries.After calcination at temperatures between520and600◦C,primary particle sizes in the range of140–220nm were achieved.Subsequent dry ball milling reduced the primary particle diameters from130to90nm,depending on time of ball milling.Reversible capacities of156mAh g−1at C/100and134mAh g−1at C/10were measured.At92%and79%of the theoretical values,respectively,these are the highest values reported to date for this material.At faster charging rates,the electrochemical performance was found to be improved when smaller LiMnPO4particles were used.©2007Elsevier B.V.All rights reserved.Keywords:LiMnPO4;Sol–gel synthesis;Lithium ion battery;Cathode material;Specific energy storage;Rate capacity and particle size effects1.IntroductionLithium transition-metal(ortho)phosphates have recently attracted attention as potential Li-ion battery cathode materi-als due to their lower toxicity,lower cost and better chemical and thermal stability,when compared to the currently used LiCoO2.The three-dimensional framework of an olivine is stabilized by the strong covalent bonds between oxygen ions and the P5+resulting in PO43−tetrahedral polyanions[1].As a consequence,olivine lithium metal phosphate materials do not undergo a structural re-arrangement during lithiation and de-lithiation.This means that they do not experience the capac-ity fade during cycling suffered by lithium transition metal oxides such as LiCoO2,LiNiO2,LiMnO2and LiMn2O4.This is attributed to structural rearrangements caused during lithiation and de-lithiation.Lithium manganese phosphate has a redox potential of4.1V versus Li+/Li[1,2],which is considered to be the maximum limit accessible to most liquid electrolytes.Unfortunately,LiMnPO4 has a low intrinsic electronic and ionic conductivity and hence∗Corresponding author.E-mail address:i.exnar@(I.Exnar).a poor discharge rate capability.The electrochemical perfor-mance is especially poor at high current densities,this is assigned to the slow lithium diffusion kinetics within the grains and the low intrinsic electronic conductivity[3,4].An approach to improve the rate performance of the olivine material is to min-imize the particle size[5,6].This reduces the diffusion path length for lithium ions in the cathode material and also creates a large contact area with conductive additives such as carbon [7–9].Delacourt et al.[10]synthesized100nm diameter particles of LiMnPO4by precipitation,which enhanced the reversible capacity to70mAh g−1at C/20from only35mAh g−1for1m diameter particles.Yonemura et al.[4]reached150mAh g−1of discharge capacity at C/100with small particles,close to the theoretical capacity of170mAh g−1.Thus,it is evident that particle size is critical in determining useful lithium capac-ity and charge/discharge rates[11–14].So far,the production of mesoparticulate LiMPO4(M=Fe,Mn)remains a challenge and only a few groups have successfully produced materials of appropriate dimensions to yield the desired electrochemi-cal performance in lithium ion batteries.The sol–gel synthesis technique offers a convenient means of producing particles of small size from homogeneous mixing of reagents on an atomic scale.In this paper we report the electrochemical performance0378-7753/$–see front matter©2007Elsevier B.V.All rights reserved. doi:10.1016/j.jpowsour.2007.06.203950T.Drezen et al./Journal of Power Sources174(2007)949–953of sol–gel prepared LiMnPO4with various particle sizes at low and high current densities.2.ExperimentalLiMnPO4powder was prepared by a sol–gel method using lithium acetate dihydrate,manganese acetate tetrahydrate and ammonium dihydrogen phosphate as precursors.The starting materials were dissolved in distilled water at room temperature and glycolic acid was used as a chelating agent.The pH of the resulting solution was adjusted to below4by the addition of con-centrated HNO3.This solution was then heated to60–75◦C to obtain a gel and the latter was dried overnight at120◦C before being heated for5h at350◦C under air.The powder as pro-duced was then calcined at different incremental temperatures (450–800◦C)for3h in air.The crystal structure of the resulting powder was exam-ined by X-ray diffraction(XRD)using Cu K␣radiation and a2θstep of0.04◦.The surface area of the samples before and after ball milling was measured by nitrogen adsorption using the Brunauer–Emmett–Teller(BET)model.Particle size distribution was measured by laser diffraction(Malvern Master-sizer).The morphology and the particle size were investigated withfield-emission scanning electron microscopy(FESEM, Philips XL30FEG).After calcination,the LiMnPO4powder was dry ball-milled(RETSCH PM4000)with20wt%of acety-lene black for4h to obtain a carbon–LiMnPO4(C–LiMnPO4) composite.Electrodes for electrochemical testing were pre-pared by tape casting a N-methyl pyrrolidone(NMP)slurry of the C–LiMnPO4composite(90wt%)with poly(vinylidene fluoride)(PVdF)binder(5wt%)and acetylene black(5wt%) on an aluminum current collector.After drying at120◦C under vacuum,the electrodes were compressed into23mm Ødisks with a thickness of27–41m,the active material loading being1.45–3.7mg cm−2.The cells were assembled in Swagelok TMfittings using Li metal foil as the counter elec-trode with a microporous polymer separator(Celgard2400TM) and liquid electrolyte mixtures containing1M LiPF6in a solvent mixture of propylene carbonate(PC),ethlylene carbon-ate(EC)and dimethyl carbonate(DMC)(1:1:3by volume). The electrochemical properties of LiMnPO4electrodes were measured by galvanostatic charge/discharge and cyclic voltam-metry using an Arbin BT2000electrochemical measurement system.3.ResultsThe crystal phase of our samples was identified from the pow-der XRD measurements to be LiMnPO4with an ordered olivine structure indexed by orthorhombic Pnmb powder prepared with an orthorhombic phase.XRD data for samples synthesized at different calcination reaction temperatures is shown in Fig.1. The olivine phase was formed at a temperature as low as350◦C even though the peak intensity was very low.The peak inten-sity increased with reaction temperature,indicating improved crystallization and particle growth.The mean particle size cal-culated from BET data,D BET(shown in Fig.2),demonstrates Fig.1.XRD diffraction patterns of the powders annealed at(a)350◦C for5h, (b)450◦C for3h,(c)520◦C for3h and(d)570◦C for3h.The᭹labeled peaks indicate an impurity phase(Li3PO4).that the average particle size increases dramatically between 600and800◦C.The powders calcinated at temperatures below 600◦C maintain a smaller D BET,of under200nm and have a well-crystallized olivine phase of LiMnPO4.The powders showed a certain degree of agglomeration with a median volume diameter D v50from the laser diffraction of around 4m.The LiMnPO4powder was dry ball-milled with20wt% of acetylene black for4h to obtain a carbon–LiMnPO4 (C–LiMnPO4)composite.During ball milling the primary par-ticle size was reduced.For a compound calcined at600◦C with an initial specific surface area of8.0m2g−1,3h of ball milling increased the BET value to18.5m2g−1(Fig.4).A particle size distribution(PSD)analysis shows an agglomeration of our mate-rial.The as-calcined LiMnPO4gave a mono-modal PSDwith Fig.2.Variation in primary particle size,D BET,of LiMnPO4as a function of annealing temperature.T.Drezen et al./Journal of Power Sources174(2007)949–953951Fig.3.Variation of the specific surface area of a LiMnPO4calcined at600◦C and volume particle size distribution(a)of the as-calcined LiMnPO4,(b)after 3h of ball milling and(c)after24h of ball milling.agglomerates of4m(D v50).PSDs of LiMnPO4with milling times from15min to24h showed bi-modal size distribution. The mean diameter decreased until3h of ball milling but then increased with further milling time(insets Fig.3).This indi-cates that the primary particles of about100nm diameter are aggregated giving a D v50of2.5m after3h milling.The pri-mary particle size is decreased by planetary ball milling but the agglomeration becomes more marked as ball milling time increases.SEM was applied to investigate the particle size and the mor-phology of materials calcinated at520,600and700◦C.As shown in Fig.4(a),the SEM images of powders calcinated at 520◦C,show small and uniform particle dimensions.In Fig.4(c) particles synthesized at600◦C are clearly larger than Fig.4(a) and the particle size distribution is broad.Reversible capacities of the C–LiMnPO4composite powders were measured at C/10and are shown in the form of cycling tests in Fig.4.The sample calcinated at520◦C demonstrated the highest capacity of134mAh g−1at C/10.The resultsdemon-Fig.5.Rate capabilities for different particle sizes of LiMnPO4. strate that the specific capacities decrease as the particle size increases.Rate capabilities for four different particle sizes are shown in Fig.5.One hundred and forty nanometer particles reached a reversible capacity of116mAh g−1at C/5(68%of the the-oretical capacity).At1C,the reversible capacities for the140 and270nm diameter LiMnPO4particles are81and7mAh g−1, respectively.Increasing particle size impacts rapidly on the abso-lute and reversible lithium capacities of LiMnPO4.This can be rationalized by a simple consideration of the lithium insertion and extraction kinetics and the poor electrical conductivity of the active phase.As the particle size increases,lithium diffusion becomes increasingly difficult due to both the diffusion limita-tion of Li+within a single large particle and the difficulty of electron transport through the bulk of thismaterial.Fig.4.The cycling performance at0.1C of LiMnPO4prepared via sol–gel using different calcination temperatures(a)520◦C,(b)570◦C,(c)600◦C,(d)600◦C and(e)700◦C and SEM images of LiMnPO4annealed at(a)520◦C,(c)600◦C and(e)700◦C.952T.Drezen et al./Journal of Power Sources174(2007)949–953Fig.6.The reversible capacities at various discharging rates in the same con-dition of charging rate of C/20.Charging and discharging rates were calculated from the nominal capacity of150mAh g−1.Fig.6shows150–160mAh g−1of reversible capacity at D/20after charging rate of C/20.Subsequently the material was charged again at C/20rate and it still obtained120–130mAh g−1 of reversible capacity at1D.At5D,65–85mAh g−1of discharge capacity was observed.The charge and discharge rate was cal-culated from a maximum practical capacity of150mAh g−1. Therefore,we reached the maximum capacity even at C/20and D/20instead of C/100and D/100.And the material achieved 87%of the maximum capacity at1D.4.DiscussionIt is suggested that to improve the electrochemical perfor-mance of olivine electrodes,it is necessary to improve both ion and electron transport.Improvement in ion transport is achieved by decreasing the particle size and ensuring a narrow size dis-tribution[13,15].Improvement in electron transport is achieved by carbon coating the cathode material and by adding carbon to the cathode.Predictably,increasing the calcination reaction temperature results in the formation of larger LiMnPO4primary particles. We demonstrate that increasing primary particle size impacts rapidly on the absolute and reversible lithium capacities of LiMnPO4.This can be rationalized by a simple consideration of the lithium insertion and extraction kinetics and the poor elec-trical conductivity of the active phase.As the primary particle size increases,lithium diffusion becomes increasingly diffi-cult due to both the diffusion limitation of Li+within a single large particle and the difficulty of electron transport through the bulk of this material.When the primary particle size is large,it will take a much longer time for lithium ions to dif-fuse into the core of the particle to make a single homogeneous phase of LiMn2+PO4or Mn3+PO4.The important observation from this work is to recognize that the mesoparticulate uniform LiMnPO4materials synthesized at lower reaction temperatures allow for shorter lithium ion diffusion lengths within a single particle.In addition,we used a high-energy planetary ball milling to make a homogeneous mixture of the active LiMnPO4mate-rial and carbon and also to make a uniform carbon network connection between carbon particles.The combination of small primary particle size and carbon coating results in enhanced specific capacities beyond92%of the theoretical capacity(156mAh g−1at C/100).The performance of116mAh g−1at C/5exceeds the best previously reported values of70mAh g−1at C/20[10]and135mAh g−1at C/25[4].5.ConclusionsNear optimal charge/discharge capacity performance has been obtained from mesoparticulate LiMnPO4cathode mate-rial in lithium ion batteries.The primary particle size for low conductivity LiMnPO4is crucial to the electrochemical performance.A relatively low reaction temperature produces well-crystallized small powders(140–160nm)with a uniform particle size and morphology.These materials achieved excel-lent charge/discharge kinetics of lithium ions.Once coated with carbon via ball milling,electrodes produced from these materials demonstrate capacities,close to the theoretical limit.The results obtained are clearly superior to literature data on LiMnPO4,con-firming that the sol–gel route provides a suitable preparation method.AcknowledgementsThe authors thank Centre Interdisciplinaire de Microscopie Electronique(CIME)for the opportunity to access SEM apparatus.We also thank te and M.Crouzet for BET measurements and electrochemical measurements,respec-tively.References[1]A.K.Padhi,K.S.Nanjundaswamy,J.B.Goodenough,J.Electrochem.Soc.144(4)(1997)1188–1194.[2]A.Yamada,S.C.Chung,J.Electrochem.Soc.148(8)(2001)A960–A967.[3]C.Delacourt,ffont,R.Bouchet,C.Wurm,J.B.Leriche,M.Mor-crette,J.M.Tarascon,C.Masquelier,J.Electrochem.Soc.152(5)(2005) A913–A921.[4]M.Yonemura,A.Yamada,Y.Takei,N.Sonoyama,R.Kanno,J.Elec-trochem.Soc.151(9)(2004)A1352–A1356.[5]A.Yamada,S.C.Chung,K.Hinokuma,J.Electrochem.Soc.148(3)(2001)A224–A229.[6]P.P.Prosini,M.Carewska,S.Scaccia,P.Wisniewski,M.Pasquali,Elec-trochim.Acta48(28)(2003)4205–4211.[7]C.H.Mi,X.B.Zhao,G.S.Cao,J.P.Tu,J.Electrochem.Soc.152(3)(2005)A483–A487.[8]S.T.Myung,S.Komaba,N.Hirosaki,H.Yashiro,N.Kumagai,Elec-trochim.Acta49(24)(2004)4213–4222.[9]H.Huang,S.C.Yin,L.F.Nazar,Electrochem.Solid-State Lett.4(10)(2001)A170–A172.[10]C.Delacourt,P.Poizot,M.Morcrette,J.M.Tarascon,C.Masquelier,Chem.Mater.16(1)(2004)93–99.[11]G.T.K.Fey,R.F.Shiu,V.Subramanian,C.L.Chen,Solid State Ionics148(3–4)(2002)291–298.T.Drezen et al./Journal of Power Sources174(2007)949–953953[12]Y.Q.Hu,M.M.Doeff,R.Kostecki,R.Finones,J.Electrochem.Soc.151(8)(2004)A1279–A1285.[13]P.P.Prosini,M.Lisi,D.Zane,M.Pasquali,Solid-State Ionics148(1–2)(2002)45–51.[14]Nam-Hee Kwon,Thierry Drezen,Ivan Exnar,Ivo Teerlinck,Motoshi Isono,Graetzel Michael,Electrochem.Solid-State Lett.9(6)(2006)A277–A280.[15]V.Srinivasan,J.Newman,J.Electrochem.Soc.151(10)(2004)A1517–A1529.。