PAH检测技术综述

A review of techniques for

the determination of polycyclic aromatic hydrocarbons in air

Sudhir Kumar Pandey,Ki-Hyun Kim,Richard J.C.Brown

We provide an extensive review of the common methodologies employed in the analysis of airborne polycyclic aromatic hydrocarbons(PAHs).The review focuses on gas-chromatography-based approaches,in the light of their universal application with excellent separation,resolution,and sensitivity.

We?rst describe collection methods for airborne PAHs in the gas and particle phases.We then evaluate the ef?ciency of extraction techniques employed for separating target PAHs from sampling media,using conventional solvent-based and emerging thermal-desorption approaches.

We also describe commonly employed analytical methods with respect to their applicability to PAHs in gas and particle phases,collected from diverse environmental settings.As an essential part of basic quality assurance,we examine each method with special emphasis on key parameters(e.g.,limit of detection and reproducibility).

Finally,we address the likely directions of methodological developments,their limitations,and the future prospects for PAH analysis.

a2011Elsevier Ltd.All rights reserved.

Keywords:Airborne polycyclic aromatic hydrocarbon;Analytical discrepancy;Analytical methodology;Extraction;Gas chromatography;Limit of detection(LOD);Polycyclic aromatic hydrocarbon(PAH);Reproducibility;Solvent-free;Thermal desorption

1.Introduction

Polycyclic aromatic hydrocarbons(PAHs) are a group of over100different chemi-cals that are known to be formed typically during incomplete combustion of organic matter at high temperatures[1].Their major sources in the atmosphere include industrial processes,vehicle exhausts, waste incineration,and domestic heating emissions,while they can also be released naturally{e.g.,forest?res[2]}.They are ubiquitous in the environment and con-tain two or more fused benzene rings in linear,angular or cluster arrangements. In addition to the frequency with which they occur in the environment,proof of their mutagenicity and carcinogenicity led to some of them being selected as priority pollutants(e.g.,16PAHs)by the US Environmental Protection Agency(EPA) (Table1).The World Health Organization (WHO)added17additional PAHs to make

a total of33PAHs under its regulation

[3].(In Table1,these33compounds are listed with their three-capital-letter acronyms).In Europe,ambient air legis-lation targets benzo[a]pyrene(with an annual target value of1ng/m3)because this compound carries the highest toxic load(de?ned as concentration multiplied by toxicity)of any airborne PAH. Because of their thermally stable struc-ture,PAHs generally exhibit a high melt-ing point,a high boiling point,and a low vapor pressure.In the atmosphere,they are distributed between gas and particle-bound phases[4–9].This phase parti-tioning is largely regulated by changes in atmospheric conditions(e.g.,temperature and relative humidity),and the physical properties of the PAHs themselves[10–13].Consequently,lighter PAHs tend to be preferentially enriched in the gas phase, while the heavier ones show almost complete association with particles[14]–indeed,in Europe,air-quality legislation to limit PAH concentrations in air sets target values for only particulate-bound PAHs (benzo[a]pyrene in the PM10particulate phase).For these reasons,it is a challenge to acquire an accurate pro?le of these

Sudhir Kumar Pandey+,

Ki-Hyun Kim*

Atmospheric Environment

Laboratory,Dept.of

Environment&Energy,Sejong

University,98Goon Ja Dong,

Gwang Jin Goo,Seoul143-747,

Republic of Korea

Richard J.C.Brown

Analytical Science Division,

National Physical Laboratory,

Hampton Road,

Teddington TW110LW,UK

*Corresponding author.

Tel.:+82234083233;

Fax:+82234084320;

E-mail:khkim@sejong.ac.kr

+Presently at:Dept.of Botany,

Guru Ghasidas Central Univer-

sity,Bilaspur(C.G.),495009,

India

17160165-9936/$-see front mattera2011Elsevier Ltd.All rights reserved.doi:10.1016/j.trac.2011.06.017

https://www.wendangku.net/doc/7b15222231.html,/locate/trac1717

harmful substances in environmental matrices,includ-ing the atmosphere.

For the analysis of PAHs in air,gas chromatography (GC)has been the most common analytical separation mechanism in combination with a number of sampling approaches and sample-preparation steps.The prefer-ence for GC over other systems[e.g.,liquid chromatog-raphy(LC)]has been demonstrated by such factors as its greater selectivity,resolution,and sensitivity[3].To improve selectivity and to ensure the robustness of

1718https://www.wendangku.net/doc/7b15222231.html,/locate/trac

sample identi?cation and quanti?cation,GC is often used in combination with mass spectrometry(MS)to produce very powerful GC-MS coupled techniques,which are able to yield even lower limits of detection(LODs)than GC alone.

This review offers a critical evaluation of the state-of-the-art information on the methodological choices available for,and recent developments in,PAH analysis in air.We discuss the procedures involved in sampling and/or preconcentration,sample preparation,and?nal determination(or detection)of PAHs in air,in order to assess and to review the nature of GC-based applications for their analysis.To this end,we initially discuss con-ventional solvent-based sample collection along with standard preparation methodologies.Moreover,we also evaluate simpli?ed solvent-free methods,as alternatives to solvent-based methods,and compare for their appli-cability in this area.Finally,we give a detailed descrip-tion of most of the essential components involved in the basic quality-assurance procedures required for the determination of PAHs in air.

2.The collection of PAHs

For the collection of airborne PAHs in both vapor and particle phases,large volumes of air must be sampled to concentrate them on the sorbent material(vapor phase) or a suitable?lter material(particulate matter).This is because their concentration in air at most locations is relatively low(of the order of ng/m3).The most common sorbent used for gas-phase enrichment is polyurethane foam(PUF).However,other sorbent materials(e.g., XAD-2,XAD-4,Carbopack C,and Tenax)have also been

employed occasionally[15–18](Table2).High-volume samplers are operated at suf?ciently high?ow rates(34–1250L/min)and for a long enough to guarantee suf?-cient sample volume for quantitative analysis(e.g.,150–900m3for gas phase and sometimes up to3000m3for particle phase).

Gas-phase samplers are usually protected from any particulate-phase material in the air sampled by the addition of a pre-?lter at the start of the sampling train. In some situations where the aim of the study is analysis of both gaseous and particulate phases,this pre-?lter may then be used for analysis of PAHs in the particulate phase,although the size-fraction characteristics of the collected particulate material will generally be unknown and unregulated(unless a size-selective inlet is em-ployed).In addition,gas-phase PAHs may be sampled using traditional PM10air samplers,where the size-fraction characteristics of the particulate phase sample are well known.In this case,the PUF cartridge sits in a tube beneath the?lter that collects the PM10particulate material.This is currently done at a limited number of monitoring sites as part of the UK PAH air quality monitoring network[19](which assists the UK to com-ply with the European Fourth Air Quality Daughter Directive,that requires Member States to monitor for B[a]P and six other PAHs).Upon completion of air sampling,these sorbents and?lter materials are treated with organic solvents(e.g.,dichloromethane,toluene, hexane,acetonitrile,and methanol)either individually or in combination to extract target PAHs.Extraction can be performed by manual washing or via automated set-ups(e.g.,Soxhlet extraction or accelerated solvent extraction).It is important to extract the PAHs from the sorbent material using as little solvent as possible in order to maximize the mass fraction of analyte in the solvent,thereby improving the overall method LOD.The ?nal determination of PAHs is then made,preferably with GC-MS.

Many studies aiming to monitor PAHs in air have attempted to measure both particle and gas-phase PAHs simultaneously(Figs.1and2).There are also a number of studies that have focused only on either one of the two,while many studies show a preference for the par-ticulate phase[20–26].As the distribution of airborne https://www.wendangku.net/doc/7b15222231.html,mon collection methods(media)for the analysis of air-borne polycyclic aromatic hydrocarbons(PAHs)in particulate phase Collection/sorbent material Target PAHs

mode a

Ref.

[A]Solvent-based extraction

(i)Glass?ber?lter(GFF)

A[29]

B[30,31,34,58]

C[25,32,54,55]

E[33]

(ii)Quartz?ber?lter(QFF)

B[37]

C[36,39]

D[38]b

E[57]

(iii)Te?on?lters

C[17,40]

(iv)Mixed?lters

XAD-4coated glass denuder B[47]

Te?on-coated glass?ber?lter E[35]

[B]Solvent free analysis(thermal desorption)

(i)QFF

A[63]

B[23,59,61] (ii)Te?on?lter B[20]

a Describes the combination of target PAHs commonly selected in many previous studies(A=some of16priority PAHs of US EPA, B=some of16+some extra,C=16PAHs only,D=16 PAHs+some extra,and E=All PAHs except16PAHs).

b Sampling with personal pumps.

https://www.wendangku.net/doc/7b15222231.html,/locate/trac1719

PAH basically occurs in both gas and particle-bound phases,all the sample-collection methods are designed to cover(1)each of the two fractions individually or(2) their sum fractions at the same time[27,28].

2.1.Collection of particle-phase PAHs

Because of complexities involved in the analysis of PAHs, some studies were simply con?ned to the particulate fractions to explore the distribution behavior of PAHs. For the collection of both particle-bound and particle plus gaseous PAHs in air,glass-?ber?lter(GFF)and quartz-?ber?lter(QFF)were used commonly(Table3). However,on rare occasions,some other materials(e.g., Te?on)were also employed individually or in combina-tion(Table3).When particulate-phase PAHs are sampled,it is usually the PM10particulate phase of ambient air that is considered,since this is currently the size fraction most relevant to human-health studies and to regulation.Note that?lter materials often used to collect particulate samples for metals analysis(e.g.,cel-lulose membranes)rarely have suf?cient strength or porosity to sustain the very high?ow rates used for sampling particulate-phase PAHs without damage or ?lter clogging.

2.1.1.Glass-?ber?lter(GFF).GFFs are the most fre-quently employed collection devices for particle-phase PAHs.The target PAHs mainly include16priority PAHs and their derivatives but are occasionally extended to additional PAHs;their sampling by GFFs has been made from diverse locations from rural to industrial areas. Although the total sample volume and?ow rate set for their collection differ between studies,most have passed suf?ciently large volumes of air through these?lter materials to ensure enrichment(for suf?cient extraction) with the common analytical methods.

Gustafson and Dickhut[29]made measurements of PAHs in air at four sites representative of rural,

https://www.wendangku.net/doc/7b15222231.html,mon collection methods(media)for vapor-phase polycyclic aromatic hydrocarbons(PAHs)in air

Collection/sorbent material Target PAHs mode a Ref. [A]Solvent based

(i)Polyurethane(PUF)

B[58]

C[29,36,40,55]b

D[31,32]

E[25,57]

(ii)XAD

XAD-4coated glass denuder B[47]

XAD-2C[17]

(iii)Combination of PUF and XAD

PUF/XAD-2/PUF C[48]

(IV)Others

Carbopack C(30mesh,400mg)c A[15]

Glass wool?lter,condenser,glass?ber?lter,Te?on?lter,and a Tenax trap d C[16]

[B]Solvent free(thermal desorber)

(i)XAD

XAD-2A[18]

(ii)Polydimethylsiloxane(PDMS)

Quartz multi-channel PDMS traps A[62]

PDMS traps C[68]

(iii)Mixed sorbents

Mixed sorbent beds((PDMS foam+PDMs particle+Tenax bed)C[49]

(IV)Others

Fan-Loy sampler(SPB-5GC columns(0.75-mm i.d.and7-l m?lm thickness)A[50]

a Describes the combination of target PAHs commonly selected in many previous studies(A=some of16priority PAHs of US EPA,B=some of 16+some extra,C=16PAHs only,D=16PAHs+some extra,and E=All the PAHs except16PAHs).

b Personal monitors.

c Diffusion-base

d sampling.

d On-line.

1722https://www.wendangku.net/doc/7b15222231.html,/locate/trac

semi-urban,urban,and industrialized areas(southern Chesapeake Bay,USA).These authors actively passed 172–365m3of air(?ow rate=510–750L/min)though GFFs to collect particle-phase PAHs.They were able to characterize14out of16priority PAHs over a wide concentration range(pg–ng/m3).

Omar et al.[30]also collected16priority PAHs in airborne PM10on GFFs from urban roadside and rural location(Kuala Lumpur,Malaysia).However,they used comparatively large sample volumes[i.e.$1612–3224m3(at?ow rate of1120L/min)]to ensure their quanti?cation at sub-ng/m3level.

In another report made by Mandalakis et al.[31],20 PAHs(16priority PAHs+Me-phenanthrenes,dimethyl-phenanthrenes,Me-chrysene,and COR)were collected on GFFs from diverse locations(an urban center,a background site,and the adjacent coastal area).These authors collected450–900-m3air samples to ensure the enrichment of the particle-bound PAHs on GFFs.

In a similar study,Tsapakis and Stephanou[32]col-lected24(16priority+8others)particle-bound PAHs by passing900m3of urban air(?ow rate=600L/min) from an urban area in Heraklion,Greece,through GFFs. GFFs have also been employed to collect a number of PAH derivatives[e.g.,oxy-PAHs and nitro-PAHs(often formed by PAH degradation–see later)],apart from16 priority PAHs.Castells et al.[33]used GFFs as the col-lection media for seven oxy-PAHs and nitro-PAHs pres-ent in urban-aerosol samples in Spain by using a very large sampling volume of1440m3(?ow rate=1000L/ min).In another study,diverse PAH components in PM (25PAHs+12nitro-PAHs+four oxy-PAHs)were quanti?ed via GFF sampling from a heavily traf?cked square in Basel,Switzerland(sample volume=960m3 at a?ow rate of nearly666L/min)[34].

On rare occasions,modi?cations of GFFs have also been used for the collection of particle-bound PAHs. Dimashki et al.[35]used Te?on-coated GFFs(often re-ferred to as Emfab)to collect particle-phase PAHs from urban atmospheres in Birmingham,UK,and Damascus, Syria.The total volume of air passed through these collection devices was900–1000m3(?ow rate=650–750L/min).

2.1.2.Quartz-?ber?lter(QFF).QFF is another preferred medium for the collection of particle-bound PAHs.Park et al.[36]collected16priority PAHs on QFFs(102mm diameter)in an urban area of Seoul(Korea)using an active sampling method.Lottmann et al.[37]collected particulate PAHs(eight priority PAHs+COR)on QFFs (150mm diameter)by drawing a total volume of 720m3(1250L/min)at a university campus (Strasbourg,France).

QFFs have also been employed to collect particle-phase PAHs with relatively low sampling volumes.Chaspoul et al.[38]attempted to collect16priority PAHs and six nitro-PAHs simultaneously from industrial workplaces at a?ow rate of1L/min on QFFs(total sample volume $0.5m3).Chantara and Sangchan[39]used QFFs with a mini-volume air sampler to collect PAHs in PM10from diverse urban areas.Note that QFFs can also be suitable as collection media for solvent-free methods[e.g.,ther-mal desorption(TD)]{e.g.,PAHs(many of the16pri-ority PAHs)and oxy-PAHs in PM2.5were collected on QFFs(sample volume=24m3at a sampling?ow rate of 1000L/min)from an urban area of Germany(Augsburg City)prior to thermal desorption GC-MS analysis[23]}.

2.1.

3.Te?on?lters and other media.Although not as common as other?ber-type?lters,Te?on?lters(e.g.,37-mm diameter)are often reported for particulate-PAH collection.As one example,these?lters were mounted in personal environmental monitors for the collection of PAHs in PM2.5(sampling duration of8h).Their sam-pling was made from indoor locations divided into smoking and non-smoking zones[40].The?ow rate of these personal sampling pumps was4L/min(with a variability of5%).

In the light of its high thermal conductivity,alumi-num foil has also been tested to sample particle-phase PAHs along with other semi-volatile organic compounds [41].The authors collected urban PM10samples on an aluminum-foil substrate installed in an Andersen eight-stage cascade impactor at a?ow rate of28.3L/min(50-h duration).Prior to use,the aluminum foils were heated to500°C for24h.The authors stated that aluminum foil has very high ef?ciency for thermal desorption,while QFF can suffer from pyrolysis of long-chain compounds during the thermal-desorption process.

2.1.4.Degradation of PAHs during sampling.It has been known for some time that,once particulate-bound PAHs become immobilized on air?lters following sampling, they become much more susceptible to degradation from oxidizing compounds,particularly ozone,but also NO2, present in the sampled air,which are drawn through the collected PM.PAHs can be oxidized upon the reaction with ozone and NO2to produce oxy-PAHs and nitro-PAHs,respectively[42].Such post-sampling reactions have been shown by Schauer et al.[43]to result in conversion of up to50%of the sampled PAHs.This,in turn,may result in a substantial underestimation of the concentrations of PAHs present in ambient air.In order to determine the quantity of degradation present in the sampled PM,the analysis of oxy-PAHs and nitro-oxy-PAHs must be undertaken which is considered more challenging[34].In addition,there is expected to be a small quantity of these compounds present in ambient air as a result of gas-phase reactions between particu-late-bound PAHs and oxidizing compounds in air.The degradation of PAHs on sampled?lters is one of the main reasons that it is prudent to limit the length of sampling

https://www.wendangku.net/doc/7b15222231.html,/locate/trac1723

periods for each PAH sample so as to minimize the effect of oxidizing gases.Indeed,the European Fourth Daugh-ter Directive[44]mandates Member States to use sam-pling periods of24h.Whilst the effect of degradation may be noticeable on short-term monitoring studies,it is thought that the effect on annual average values(e.g., those produced by?xed air-quality networks)is likely to be small.As the annual variations in the concentration of PAHs and ozone are in anti-phase,there is very little effect on measured PAH concentrations.For example,in the winter(e.g.,increased fuel use),PAH concentrations become high whilst ozone concentration is low.Con-versely,in the summer when the ozone is high,PAH concentrations are low.Hence,although relative deg-radation rates can rise,these have little effect on the annual average.Moreover,judicious timing of the automatic?lter-changeover facility on many of the current air-quality-monitoring samples can allow?lter changes to occur at dawn,to take advantage of the diurnal variation in ozone concentrations and to achieve a large period of time during the night prior to sample changeover when ozone is low.Thus,the degradation to PAHs collected during the day,and subsequently over-night,is minimized.A full understanding of the role of ozone in PAH degradation is still being developed[45]. Recent developments to solve the ozone-degradation problem have included use of an ozone denuder in combination with the PM10sampler[46].This innova-tion removes ozone from the sampled air and prevents degradation of PAHs within the sampled PM.However, these devices may be less effective in humid and wet conditions,so the same group is currently working on an internal heater for the ozone scrubber to solve this problem.

2.2.Collection of gas-phase PAHs

If we review the literature available for airborne-PAH analysis,the dominant portion is found to cover16 target PAHs(of the US EPA)in both particulate and vapor phases.As the collection devices for particle-phase PAHs have been aforementioned,we now focus on the substances used for the collection of gas-phase PAHs in the following sub-sections.Because of the low concen-trations of PAHs present in air,their collection usually relies on pumped sampling(where a known volume of air is drawn through the sorbent)rather than diffusive sampling(where vapor is allowed to diffuse into the sorbent).This is because diffusive samplers are in many times not simple to calibrate for quantitative analysis.

2.2.1.Polyurethane foam(PUF).PUF has been utilized as the most common means of enriching gas-phase PAHs.It is applicable for a wide range of PAHs(16 priority PAHs and many others,especially their nitro and oxy derivatives).Gustafson and Dickhut[29]col-lected gas-phase PAHs(14out of16priority PAHs)on two PUF plugs by passing172–365m3of air(?ow rate=510–750L/min)at both rural and industrial sites.Mandalakis et al.[31]used a PUF plug (length=8.0cm,diameter=7.5cm)to enrich20PAHs (16priority PAHs+four others)by passing450m3and 900m3of air from diverse locations(an urban center,a background site,and the adjacent coastal area).In a similar study,Tsapakis and Stephanou[32]collected24 airborne PAHs(16priority PAHs+eight others)by passing900m3of urban air(?ow rate=600L/min) through PUF.

The applicability of PUF samplers has also been dem-onstrated in the determination of diverse PAH deriva-tives(e.g.,nitro-PAHs and oxy-PAHs)in the vapor phase.Dimashki et al.[35]used PUF(80mm diameter, 75mm depth,density=0.02g/cm3)to collect vapor-phase PAHs;they used a sampling volume of900–1000m3at a?ow rate of650–750L/min from two urban locations(Birmingham,UK,and Damascus, Syria).

2.2.2.Collection on XAD.As an alternative to PUF, Possanzini et al.[47]tested an XAD-4(20–60mesh) coated-glass denuder(with QFF for particle)for the col-lection of gas-phase PAHs.They assessed collection ef?-ciency(and capacity)over6-h sampling at an air?ow rate of6L/min.The collection ef?ciencies of two-ring and three-ring PAHs were found to exceed90%[e.g., 90%(PYR)–97%(NAP)].

2.2.

3.Mixed sorbents/others.For the collection of gas-phase PAHs,PUF samplers have been used most inten-sively in combination with other sorbents(e.g.,XAD-2). Lee et al.[48]used a glass cartridge containing a5-cm PUF plug,followed by3-cm XAD-2resin,and?nally a2-cm PUF plug to collect the gas-phase PAHs from an area affected by traf?c activities.Bertoni et al.[15]proposed a molecular diffusion-based sampling device for the determination of gaseous PAHs in air.These authors used Carbopack C as the adsorbing material to enrich the PAHs for the GC-MS analysis after two months?exposure of the sorbent in urban or suburban air.

Wauters et al.[49]used a mixed sorbent bed{poly-dimethylsiloxane(PDMS)foam(1cm),PDMS particles (120mg),and a Tenax TA(60mg)}for the collection of 16priority PAHs by a personal air pump at a?ow rate of 100mL/min for24h.They compared their method to the classical method relying on high-volume sampling onto a PUF sampler for24h.The total volumes of samples collected by these methods contrasted between 144L(mixed bed)and1296m3(classic method). However,they found the new approach superior to the conventional method,as the new approach yielded much higher recoveries–ranging from1.2times better for PHN to35times better for NAP–despite the much smaller sampling volume.

1724https://www.wendangku.net/doc/7b15222231.html,/locate/trac

An online sampling system was also designed to col-lect the16priority PAHs emitted from a coal-?red?u-idized bed combustor(sample volume=60L at a?ow rate of2.5L/min).This system was built to combine a glass-wool?lter(to remove particles),condenser(to remove water),GFF,Te?on?lter,and a Tenax trap(as the trapping device for PAHs)[16].In addition,for assessing personal exposure to PAHs in urban-community settings,Fan et al.[50]prepared a honey-comb-like sampler with320sections of1-cm long SPB-5 GC columns(0.75-mm i.d.and7-l m?lm thickness)and collected gaseous PAHs(NAP,ACN,ACL,FLR,PHN, ANT,FLT,and PYR)on the inner surfaces of the col-umns by molecular diffusion.Their method was suf?-cient to enrich these gas-phase PAHs for further analysis with TD-GC-MS.

3.Evaluation of recoveries for different extraction techniques

As the analysis of PAHs generally relies on collection media for pre-concentration,regardless of their phase, one needs to know how effectively their extraction pro-ceeds in the sampling stage.It has always been a chal-lenge to extract all the target PAHs with suf?ciently good ef?ciency due to large variations in their physicochemi-cal properties(e.g.,volatility and reactivity).The use of surrogate standards directly spiked to the sorbent or?lter material before sampling(or after sampling if the sam-ples are measured?as received?and losses during sam-pling are not considered)is usually recommended for recovery calculations.For example,the best surrogate standards are isotopically-labeled congeners(e.g.,deu-terated analogues of the PAHs being measured)[3].It is important to bear in mind that,whilst isotopically-labeled PAH congeners are expected to mirror almost exactly the physical and chemical properties of their non-isotopically labeled counterparts,because they are spiked onto collected PAHs,they can never exactly replicate the matrix of the as-collected sample.This is especially true for PAHs bound to particulate matter.The recovery of isotopically labeled congeners may,therefore, be slightly higher than that for the sampled PAHs–although the inherent variability of recovery data makes it very dif?cult to draw such conclusions with con?dence. Although many studies dealing with PAH determination do not provide the full details of recovery data,numerous efforts have been directed to the estimation of extraction ef?ciencies[51,52].The results of such efforts can be examined for both the gas and particle phase.

3.1.Recoveries of solvent-based extraction for particle-phase PAHs

After the collection of airborne particulate PAHs,the ?lter samples are treated with a range of organic solvents (e.g.,hexane,dichloromethane,acetone and methanol) individually or in combination.Moreover,to enhance the recovery of PAHs from the?lter materials,solvent-based extraction is done by specialized apparatus(e.g., Soxhlet,microwave extraction,or ultrasonication)or other techniques(e.g.,pressurized?uid extraction) [53,54].

Gustafson and Dickhut(28)extracted PAHs from GFF samples by combining acetone,petroleum ether,and dichloromethane as extractants in a Soxhlet apparatus. The recoveries of surrogate PAHs(i.e.Anth-d10,BaA-d12,and BaP-d12)were92.6±22.1%,103.1±9.5%, and79.3±13.5%,respectively.

Omar et al.[30]estimated recovery of particle-phase PAHs on GFFs with ultrasonication in dichloromethane as50.8–99.7%.However,these authors also reported that the lighter PAHs(NAP,ACN,ACL,and FLR)could not be accurately identi?ed by this method due to low recoveries and poor reproducibilities.

By contrast,Zhang et al.[55]extracted particle-phase PAHs on GFF with hexane and cyclohexane and ob-served superior recovery rates of84.5%(NAP)to108% (BAP).

The European Committee for Standardization(CEN) Working Group(TC264WG21),which developed and wrote the reference method for benzo[a]pyrene mea-surement in air(the PAH subject to target-value assessment in Europe),found relatively little difference in performance between a number of solvents tested for their extraction ability.Indeed,most solvents and sol-vent combinations are allowed by the standard method, as long as they achieve satisfactory recoveries when analyzing NIST Standard Reference Material1649a. However,their standard does make the following rec-ommendations for‘‘preferred solvents’’,depending on extraction technique[56]:

(1)Extraction under re?ux:toluene

(2)Soxhlet extraction:toluene,dichloromethane,1:1

hexane:acetone mixture

(3)Microwave extraction:1:1hexane:acetone mixture

(4)Accelerated solvent extraction:toluene,dichlorometh-

ane,1:1hexane:dichloromethane mixture

(5)Ultrasonic extraction:toluene,dichloromethane

As summarized in Table2,many researchers have attempted to describe the extraction ef?ciency of priority PAHs.Extraction ef?ciencies may also be adversely af-fected if solid-phase extraction(SPE)and/or sample clean-up procedures are required prior to analysis.These are regularly required for ambient air samples because of frequent instances of carbonaceous and particulate contamination after extraction.However,unlike those common target PAHs,it has been an even greater challenge to acquire optimum recoveries for their derivatives(e.g.,oxy-PAHs and nitro-PAHs)[34].For instance,nitro-PAHs and oxy-PAHs from GFF?lters were extracted in methanol/acetone with the assistance

https://www.wendangku.net/doc/7b15222231.html,/locate/trac1725

of a microwave[34].They were further puri?ed with SPE and analyzed by semi-preparative high-performance liquid chromatography(HPLC).The GFF recovery rates of nitro-PAHs and oxy-PAHs were in the range80–90%. Dimashki et al.[35]found relatively good recovery rates of particle-bound PAHs on GFFs in the range61–98%, when extracted via Soxhlet with dichloromethane. Albinet et al.[57]used QFFs for collecting nitro-PAHs and oxy-PAHs in the particulate phase.They extracted all the samples in dichloromethane and proceeded with an LC puri?cation procedure on an SPE cartridge.The recovery rates of nitro-PAHs in the whole analytical procedure were estimated as14%(1-nitronaphthalene) to84%(7-nitrobenz[a]anthracene),while those of oxy-PAHs were from5%(1-naphthaldehyde)to83% (benzo[b]?uorenone).

It is usual practice in particulate-based PAH studies to correct the?nal results based on the recovery of the surrogate standards spiked into the real samples.Where there is a direct isotopic analogue measurement avail-able for the target PAH in question,this is simple.Where there is no direct isotopic analogue measurement avail-able for the target PAH in question,expert judgment needs to be used to choose the most appropriate isoto-pically-labeled PAH for performing the recovery correc-tion.This is likely to be most chemically similar and closely-eluting isotopically-labeled PAH from the gas chromatogram.

3.2.Solvent-based recoveries of gas-phase PAHs

As for?lter materials for collecting particulate PAHs, sorbent materials for collecting gaseous PAHs can also be subject to extraction by organic solvents with the modi?cation of various treatment techniques.Gustafson and Dickhut[29]estimated recoveries of surrogate PAHs (i.e.ANT-d10,BAA-d12,and BAP-d12)in PUF plugs. These authors found recovery rates of89.3±18.5%, 100±20.7%,and87.2±19.8%,respectively,based on Soxhlet treatment with acetone,petroleum ether,and dichloromethane(as extractants).Simcik et al.[58] achieved recovery rates of78–101%from surrogate standards treated on PUF after extraction with dichlo-romethane.Zhang et al.[55]collected gas-phase PAHs on PUF plugs and then extracted them with hexane and cyclohexane.They were able to yield recovery rates of 63.2%(Nap)to108%(BaP).Supercritical?uid extrac-tion(SFE)with dichloromethane(plus ultrasonication) was employed to extract gas-phase PAHs from PUF[36]. The recovery rates of this application were69–95% (mean=83%).

For the analysis of gas-phase nitro-PAHs(from PUF), Dimashki et al.[35]employed Soxhlet apparatus in dichloromethane and found recovery rates of57–92%. By contrast,the recovery of PAHs from Carbopack C ranged from53%(NAP)to100%(CHY),when they were extracted by a single solvent(toluene)at ambient temperature[15].

According to the above comparison,the recovery rates of gaseous PAHs vary over a wide range,depending upon both the types of extraction technique and the individual PAHs.However,compounds of low molecular mass with relatively high volatility(e.g.,NAP)tend to show the lowest recoveries,regardless of phases(gas and particle).

3.3.Extraction ef?ciency of solvent-free or thermal-desorption methods

In recent years,many efforts were made to?nd an alternative analytical procedure for PAHs.As most of the previous analyses generally relied on the classic,solvent-based approach,more recent studies were directed to-ward use of solvent-free methods(e.g.,TD systems). Datasets on the extraction ef?ciency of these solvent-free methods are still scarce.In most cases,the reliability of such techniques cannot be assessed directly against those of conventional solvent-based methods,regardless of PAH-phase type.

3.3.1.Particle phase.Drooge et al.[59]evaluated recovery ratios of some particle-bound PAHs collected via QFFs based on comparative analysis between the standardized liquid extraction(LE)method and the TD method.In case of PHN and ANT,they found recovery ratios in the TD method higher than those in the LE method.However,the recovery of other PAHs in the TD method was comparable to those in the LE method.Bates et al.[60]were able to achieve their TD recoveries in the range95.5%(BGP)to99.1%(BAA)for samples collected on QFFs.In other research,Gil-Molto et al.[61]esti-mated the recovery ratios of12PAHs(NIST SRM1649a urban dust)on QFFs through comparative analysis by the TD method.They found the recovery ratios of par-ticulate PAHs in the range95.01%(ICP)to99.59% (PHN).

3.3.2.Gas phase.In solvent-based methods,extractions of gaseous PAHs have been conducted almost solely from PUF.However,the solvent-free techniques(i.e.the TD method)are suf?ciently advanced to allow comparison between diverse sorbent materials.Wei et al.[18]used a microwave-assisted desorption(MAD)coupled to head-space solid-phase microextraction(HS-SPME)as a one-step sample-preparation method for PAHs with XAD-2 adsorbent.Although their method cannot be regarded as fully solvent-free,they were able to exclude the use of any toxic organic solvent.The eight target PAHs(out of 16priority PAHs)on XAD-2were then desorbed into ethylene glycol and evaporated into the headspace via microwave irradiation.These PAHs were then absorbed directly on an SPME?ber in the headspace for?nal

1726https://www.wendangku.net/doc/7b15222231.html,/locate/trac

detection by MS.Their study achieved a recovery rate of >80%under optimized extraction conditions.

Fan et al.[50]estimated>85%recovery of eight gaseous PAHs(NAP,ACN,ACL,FLR,PHN,ANT,FLT, and PYR)from a honeycomb-like passive sampler(320 sections of1-cm long SPB-5GC columns)by a TD method.However,large variabilities were observed in the recoveries of FLT and PYR with relative standard deviations(RSDs)of36%and30%,respectively.These unstable trends were suspected to come from tempera-ture inhomogeneity in the thermal-desorption system. Wauters et al.[49]found the recovery of16priority PAHs in the range of80.5%(NAP)to118%(BGP) through combined application of a mixed sorbent-bed and solvent-free TD-GC-MS analysis.

4.The solvent-based analysis of PAHs

Most of the previous studies aiming to investigate the distribution of airborne PAHs(particle or gas phase) have relied on solvent-based analysis,once the initial collection of PAHs is completed on a suitable?lter or sorbent material.The PAHs bound on those media are then extracted with organic solvents(in liquid phase)for the?nal determination,preferably with GC-MS.These conventional methods have been employed successfully to describe the distribution pattern,source apportion-ment,and phase distribution(gas or liquid)of PAHs from diverse environmental settings(rural,urban, industrial,indoor,workplace,and laboratory condi-tions).In this section,we discuss the common and modi?ed approaches adopted in the solvent-based determination of PAH concentrations,based on the key phase-classi?cation criteria(i.e.particle and gas),as we did for sample-collection methods(Tables4and5).

4.1.Solvent-based analysis of particle-phase PAHs Gustafson and Dickhut[29]were able to analyze14out of16priority PAHs using samples collected from diverse locations around southern Chesapeake Bay,USA.These authors conducted GC-MS analysis on particle-phase PAHs collected on GFFs after extraction in dichloro-methane with the aid of ultrasonication.The GC-MS-based determination yielded PAHs in a concentration range of0.58pg/m3(DBA at a rural site)to0.86ng/m3 (CHY in an industrial area).Omar et al.[30]reported BGP and COR as the most abundant PAHs in airborne PM10from urban roadside and rural location(Kuala Lumpur,Malaysia)based on the GC-MS analysis after solvent-based extraction of PAHs on GFFs. Concentrations of16priority PAHs bound with PM10 samples(from diverse urban areas)were determined by a GC-MS method after extraction in acetonitrile with an ultrasonicator[39].The MS was operated in single-ion monitoring(SIM)mode to quantify characteristic ions of selected PAHs with the aid of the deuterated internal standard(closest to the analyte in molecular weight). They were able to report the mean concentrations of total PAH(sum of16)in the range2.7ng/m3(rural area)to16.6ng/m3(traf?c area).

In another study,Park et al.[36]analyzed16priority PAHs in the concentration range0.14ng/m3(ACL)to 4.89(BBF)ng/m3,using GC-MS-based method after collection on QFFs and extraction in dichloromethane with ultrasonication.They used external PAH standards (16PAHs in a mixture)to quantify the individual ana-lytes by GC-MS analysis.

Mandalakis et al.[31]monitored20PAHs(16priority PAHs+Me-phenanthrenes,dimethyl-phenanthrenes, Me-chrysene,and COR)in samples collected from diverse locations(an urban center,an inland site and a coastal area).They analyzed the target PAHs in concentration range of0.01ng/m3(Me-Chrysene in the inland area)to 7.48ng/m3(PHN in the coastal area)by GC-MS analysis in SIM mode after extracting the?lter(GFFs)in hexane (Soxhlet extraction).

In a similar study,Tsapakis and Stephanou[32] analyzed24particle-bound PAHs from an urban area of Heraklion,Greece.Extraction of PAHs collected on GFFs was done with hexane in Soxhlet apparatus,and24 PAHs(16priority PAHs+8additional ones)were quanti?ed in a concentration range of0.02ng/m3(FLR) to3.27ng/m3(BGP)by GC-MS-based analysis in SIM mode.

Numerous modi?cations have also been introduced in the analytical procedures of solvent-based methods. Chaspoul et al.[38]attempted to analyze16priority PAHs and six nitro-PAHs simultaneously from industrial workplaces by GC-MS analysis following solvent-extraction(QFF samples).They used two different ionization modes in MS[i.e.electron-impact(EI)for priority PAHs and electron-capture negative ionization (ECNI)for nitro-PAHs].These authors were able to analyze the nitro-PAHs and priority PAHs in the ranges 0–20414ng/m3and65–22180ng/m3,respectively. Personal environmental monitors mounted with Te?on?lters were also used for particulate PAHs anal-ysis from indoor locations divided into smoking and non-smoking zones in Taiwan[40].These authors could quantify airborne PAHs as the geometric mean of all16 priority PAHs in particle phase(460ng/m3)through a GC-FID-based determination.

The analysis of nitro-PAHs and oxy-PAHs was accomplished by high-resolution GC combined with ion-trap HRGC-MS2[34].This method was effective enough to monitor nitro-PAHs in the range2–62pg/m3,which was10–100times lower than their oxy-PAH counter-parts.

A two-step supercritical-?uid extraction(SFE)method was also developed for the analysis of oxy-and nitro-PAHs present in urban aerosol samples(in Barcelona,

https://www.wendangku.net/doc/7b15222231.html,/locate/trac1727

Spain)by GC with electron-capture detection(ECD) coupled to MS[33].The SFE method,applied to the analysis of oxy-and nitro-PAHs in urban aerosol sam-ples,was able to identify9-?uorenone,9,10-anthraqui-none,2-methyl-9,10-anthraquinone,benzanthrone, benz[a]anthracene-7,12-23dione and1-nitropyrene at a concentration range of15–364pg/m3.

Dimashki et al.[35]used Te?on-coated GFFs to collect particle-phase PAHs from Birmingham,UK,and Damascus,Syria.These authors conducted the Soxhlet extraction in dichloromethane for subsequent GC-MS analysis and found nitro-PAHs in the range0.13–0.42ng/m3.

4.2.Solvent-based analysis of gas-phase PAHs

Similar to the methods employed in the analysis of par-ticle-phase PAHs,those developed for gas-phase PAHs have also relied on extraction of the sorbent material. Hence,gas-phase PAHs(e.g.,PUF)extracted into some organic solvents can be subject to?nal determination by GC-MS.

Gustafson and Dickhut[29]measured gas-phase PAHs at four sites representative of rural,semi-urban, urban,and industrialized areas.They collected gas-phase PAHs on two PUF plugs.Those samples were extracted in organic solvents(e.g.,acetone,petroleum ether,and dichloromethane)with Soxhlet apparatus and analyzed for14PAHs with deuterated PAH-surrogate standards. The extracts were then concentrated using rotary evaporation followed by evaporation under puri?ed nitrogen.These samples were again extracted with hexane and cleaned using solid-liquid chromatography on silica gel for the?nal determination using GC-MS. They characterized14out of16priority PAHs in a concentration range0.58pg/m3(BAP at rural site)to 124ng/m3(PHN)in the urban area.

In another study,Park et al.[36]collected16priority PAHs in gas phase using a PUF sampler in an urban area of Seoul,Korea.The analysis by a GC-MS method(in SIM mode)after extraction with dichloromethane(with ultrasonication)revealed gas-phase PAHs falling in the range0.80ng/m3(BAA)to15.03ng/m3(PHN). Mandalakis et al.[31]monitored20PAHs(16priority PAHs+four others)in samples collected from diverse locations(urban center,an inland site and the adjacent coastal area).They could determine target PAHs in the range0.01(PHN)(Me-CHY in the inland area)to 7.48ng/m3(PHN in the coastal area)by combining extraction with hexane(in Soxhlet apparatus)and GC-MS analysis.

In a similar study,Tsapakis and Stephanou[32]col-lected samples of24PAHs(16priority PAHs+eight others)in urban air on PUF,extracted them with hexane in Soxhlet apparatus,and analyzed them by GC-MS in SIM mode.They determined them in the range 0.001ng/m3(COR)to19.8ng/m3(PHN).

Bertoni et al.[15]also analyzed gas-phase PAHs in concentration range231–1743ng/m3(NAP),11–51ng/m3(PHN), 2.1–12.1ng/m3(FLT),and9.2–39ng/m3(CHY)in an urban area of Rome,Italy,with a combination of a passive sampling technique(Carbopack C as sorbent)and a GC-MS method.

Dimashki et al.(35)analyzed nitro-PAHs in the gas phase by treatment of a PUF sample in dichloromethane with Soxhlet apparatus.They determined them in the range0.01ng/m3(9-nitro-ANT)to0.21ng/m3(1-ni-tro-NAP)in an urban area of Birmingham,UK,by a GC-MS analysis in NICI mode.

5.Solvent-free methods for PAH analysis

Solvent-based extraction methods,if employed in the determination of atmospheric PAHs,generally involve time-consuming,labor-intensive procedures,regardless of PAH-phase type.As this step requires the use of toxic organic solvents,it may cause added dif?culties with sample handling during the extraction stage[62].To overcome such limitations,solvent-free methods(e.g., TD)have commonly been employed.Their applicability has been demonstrated in the analysis of VOCs and re-duced sulfur compounds(RSCs)[63].As for solvent-based methods,the applicability of this technique may also be considered separately for gas-phase and particle-phase PAHs.

5.1.Solvent-free analysis of particle-phase PAHs

Two types of?lters(i.e.QFF and Te?on)have commonly been utilized for the collection of the particle-bound PAHs that are subsequently retrieved directly without using solvent.

As an example of the use of a Te?on?lter for solvent-free collection device,Bezebeh et al.[64]analyzed aerosol-particulate matter for some of the16priority PAHs and nitro-derivatives using laser-desorption-ionization time-of-?ight MS(LDI-TOF-MS).They col-lected aerosol particulate matter(PM2.5)on Te?on?lters from inside a bus terminal.The analytes were introduced into the LDI-TOF-MS by mounting a3-mm diameter piece of the?lter on a probe.The probe was rotated to allow the analysis of samples at multiple positions on the aerosol?lter.The collection of a small sample volume (0.32–0.98m3)and desorption and ionization with pulsed UV radiation at266nm resulted in detection of PAHs in the positive-ion spectra,while nitro-PAHs were detected in the negative-ion spectra.By applying this technique,they were able to quantify several priority PAHs(PYR,CHY,TPN,PRL,and DBA)and three nitro-PAHs at nearly10ng/m3and0.1ng/m3,respectively. QFFs have been employed frequently to collect par-ticle-bound PAHs for subsequent solvent-free analysis. PM2.5samples collected on QFFs from an urban area of

1730https://www.wendangku.net/doc/7b15222231.html,/locate/trac

Germany(Augsburg city)were analyzed for many of the16priority PAHs and oxy-PAHs by direct thermal desorption(DTD)-GC-TOF-MS[23].For this purpose, the?lter samples were cut into pieces(2.1·27mm). Two?lter sub-samples,each of which represented 1.1m3of sampled air,were placed into a GC-liner. Isotope-labeled reference compounds were added(for quanti?cation),and liners were put into the cold injector.Desorption of target PAHs was accomplished at320°C(for15min).Based on this method,these authors quanti?ed the target PAHs in the sub-ng/m3 range.

In a similar study,a two-step TD injection system integrated into a GC-MS was applied as a solvent-free method for the determination of PAHs across a wide volatility range(PHN to BGP)from ambient air particles [59].These authors collected PM10samples on QFFs at high volume(720m3at1250L/min)and low volume (6.9m3at38.3L/min)from an urban location in Wiesbaden,Germany.These QFF samples were then cut into pieces,and one piece(corresponding to a high sample volume of180m3)was placed in a glass liner. After adding an internal standard,the?lter piece was thermally desorbed in a TD unit interfaced with a GC-MS system.These authors determined PAHs in the concen-tration ranges of0.03–2.36ng/m3(BAP)and0.03–4.86ng/m3(Retene).

Likewise,Gil-Molto et al.[61]relied on a TD method to analyze12(out of16)priority PAHs on QFFs(in PM10 and PM2.5samples),collected by low-volume samplers from the city of Elche,Spain.These?lter samples were rolled and placed in TD tubes to desorb the target PAHs at300°C(for10min)in an automatic TD system interfaced with GC-MS.These authors successfully determined the PAHs in both PM2.5and PM10samples in the concentration ranges0.020ng/m3(DBA)to0.41 ng/m3(CHY)and0.002ng/m3(ANT)to0.94ng/m3 (BBF),respectively.

The reliability of the TD-GC-MS technique was eval-uated for4–6ring PAHs collected on PM10quartz?lters [60].Their TD device was equipped with a desorption oven connected to a Peltier-cooled,sorbent-packed,cold-trapping system.Through application of a two-stage focusing technique,they were able to achieve complete recovery of samples of a NIST urban dust[Standard Reference Material(NIST SRM1649a)]at the95% con?dence level based on the certi?ed values supplied (except for BAA).

For the speciation of organic components,including PAHs in ambient air,PM2.5samples were analyzed by a TD-GC-MS method[65].The authors collected PM2.5 samples on QFFs and cut two separate punches(8.1mm in diameter,0.518cm2in area)to load into the TD glass tube.The samples were then analyzed via thermal desorption combined with GC-MS.Based on this method, these authors quanti?ed the PAHs in PM2.5samples in the range0.02ng/m3(FLR)to1.64ng/m3(ICP)from an urban area(Golden BC,Canada).

Falkovich and Rudich[41]reported a direct sample-introduction(DSI)method for?lter or liquid samples into a GC injector for thermal desorption and subsequent GC-MS analysis to identify a number of semi-VOCs,includ-ing16priority PAHs.Their sampling was conducted on aluminum foil(preferred over QFFs by these authors)to be desorbed in micro-vials inside the GC injector(tem-perature400–450°C).For quanti?cation of target PAHs, 2l L of the PAH standard solutions(EPA610mixture) contained in micro-vials were thermally desorbed and similarly analyzed on aluminum foil.In this way,they were able to quantify PAHs in the concentration range 0.005ng/m3(PYR)to0.34ng/m3(PHN)in PM10 samples collected from an urban area in Tel Aviv,Israel. In an effort to improve a solvent-free quanti?cation method for particle-phase PAHs,Waterman et al.[66] conducted a TD-GC-MS analysis of NIST SRM1649a (urban dust).They prepared samples with a known quantity of urban dust in a glass-lined,stainless-steel GC liner(cleaned at600°C for5min)for thermal desorp-tion.The samples vaporized from the liner were then focused on a cold trap(à196°C with liquid nitrogen) and?ash heated fromà196to300°C for20s for sub-sequent GC-MS analysis.They were able to con?rm that the values that they obtained agreed with the certi?ed values at the95%con?dence interval.

As a solvent-free technique,direct HS-SPME(with a PDMS?ber)was also developed for the determination of particulate PAHs collected on Te?on-coated GFF[67]. HS-SPME extractions were performed in0.6mL of20% sodium sulfate in deionized water.The extraction was conducted in two steps:(1)equilibration of the?lter (with PAH sample)and the liquid phase(sodium sulfate in water)and(2)SPME extraction at50°C for120min. These authors were able to quantify the16priority PAHs in the concentration range of0.20(ANT)to 1.8ng/m3(PYR)in PM10samples collected from an urban area in Brazil(Sao Paulo city)by GC-MS analysis.

5.2.Solvent-free analysis of gas-phase PAHs

For solvent-free analysis of gas-phase PAHs,many types of sorbent materials have been tested and evaluated, either individually or in combinations.Baltussen et al.

[68]proposed an analytical method for gas-phase PAH analysis based on enrichment of the solutes on a packed bed of100%PDMS particles,followed by TD-GC-MS.The PDMS material exhibited excellent thermal stability even after200consecutive https://www.wendangku.net/doc/7b15222231.html,ing the above method, they were able to determine16priority PAHs in the concentration range of0.50ng/m3(BAP)to129ng/m3 (ACL).

Recently,a novel analytical method for atmospheric PAHs was developed based on laser-induced?uorescence (LIF)of samples on quartz multi-channel PDMS traps

https://www.wendangku.net/doc/7b15222231.html,/locate/trac1731

with the aid of diffusion-based sampling.The method allowed the rapid(<5min),cost-effective analysis of samples[62].However,small amounts of naphthalene-photodegradation products(phenol,benzyl alcohol,and phthalic anhydride)were also identi?ed by TD-GC-MS after>15min irradiation.

MAD coupled to headspace solid-phase micro-extraction(HS-SPME)has been studied for in-situ,one-step,sample-preparation step for PAHs collected on XAD-2adsorbent for subsequent GC-MS analysis[18]. The PAHs on XAD-2were desorbed into the extraction solution(ethylene glycol),evaporated into the headspace by microwave irradiation,and absorbed directly on a SPME?ber in the headspace.PAHs were then analyzed by desorbing the?ber in the injection port.This method was applied to collect smoke samples from the indoor burning of joss sticks[18].The quantities of PAH determined via this approach were in the range0.795–2.53ng.

Wauters et al.[49]used a mixed sorbent bed(i.e. PDMS foam,PDMS particles,and a Tenax TA bed)for the collection of16priority PAHs in the gas phase. Quantitative analysis of those samples was then made by TD-GC-MS.They also compared their method to the classical method that typically relies on the high-volume sampling(on GFF followed by PUF)for24h.These au-thors could monitor the16priority PAHs in the con-centration range0.45ng/m3(DBA)to116ng/m3 (NAP)from the sample collected inside the university campus at Ghent,Belgium.

A sensitive,solvent-free method was also developed to quantify personal exposure to gaseous PAHs(NAP,ACN, ACL FLR,PHN,ANT,FLT,and PYR)via passive sam-pling with a lab-made sampler[50].They were then able to analyze gas-phase PAHs in the concentration range 1.9–2600ng/m3by a TD-GC-MS-based analysis.

6.Evaluation of basic quality assurance in PAH analysis

Considering that airborne PAHs are present at very low concentrations(at or below the ng/m3range),the most important parameter for the basic quality assurance (QA)is LOD,which is often expressed in terms of abso-lute mass(e.g.,ng)and/or mass per volume(e.g., ng/m3).Although most of the studies dealing with PAH analysis relied on GC-MS,the magnitude of LOD values varied noticeably,depending upon the analytical con-ditions and instrumental set-ups.Moreover,it also varied between the PAH phases(e.g.,gas or particle).It is also an important consideration that,because GC-MS is a highly speci?c technique,whereby an analyte is only detected if it elutes from the GC at the right time,and,in addition,exhibits the molecular ion fragment being monitored in MS,LODs based on the standard formula of three times the standard deviation of the blank responses are sometimes very dif?cult to calculate.On occasions where the blank response is too low,other methods to calculate the LODs based on?t-for-purpose uncertainty assessment,and sample adulteration with impurities have been used[69].If the LOD values are to be ex-pressed in concentration terms to yield a method LOD, the volume of total samples is also a critical variable to assess.In addition,the LOD will be affected by many other method parameters(e.g.,the reduced volume of the?nal extract and the volume injected into the GC-MS) [70].

To provide an overview of the normal detection range of PAHs across different analytical approaches,the LOD values of different studies are compared in Table6 (particle)and Table7(vapor phase).As shown in Table6,the LOD values for particle-phase PAHs varied over the ng to sub-ng range.Although these LOD values are also variable across individual PAHs,the data for each PAH are rather scarce.In addition,most studies reported a range of LOD values for their analysis instead of each individual one.

Simcik et al.[58]calculated LOD values for10target PAHs(seven of16priority+three extra)in the particle-phase and reported their LOD range as0.002–59ng based on GC-MS analysis(after solvent treatment). Comparatively high LOD values were often reported for some lighter PAHs with high volatility(e.g.,NAP). Possanzini et al.[47]reported an LOD value of5.9ng for NAP in the particulate phase,which was an order of magnitude greater than their ANT value(0.3ng);these authors also relied on a solvent-based GC-MS analysis. Although a wide variation in LOD values was reported for many individual PAHs,some authors also found a relatively narrow range of LODs across different indi-vidual PAHs(Table6).Van Drooge et al.[59]could achieve a limit of quanti?cation(LOQ)value near 0.08ng for most of the16priority PAHs in particle phase,based on a two-step TD and GC-MS method.In a similar study,Gil-Molto et al.[61]also found their LOD values in the sub-ng range(0.01–0.5ng)for16priority PAHs in particle phase by TD-GC-MS analysis. According to our assessment of a large number of LOD values,the values for vapor-phase PAHs generally fell near,or below,the ng range.Lee et al.[48]reported LOD values for16priority PAHs in the range0.07–0.15ng with a solvent-based GC-MS analysis.In another study, Wei et al.[18]reported LOD values in the range0.02–1ng using an HS-SPME-GC-MS method.Possanzini et al.

[47]also found relatively high LOD values in vapor phase[i.e.1ng(CHY)to15.3ng(NAP)by a GC-MS method(solvent-based)].Likewise,Simcik et al.[58] found a comparatively wide range of LODs for vapor-phase PAHs(i.e.0.002–1406ng)in a solvent-based GC-MS analysis.Apart from these PAHs,Albinet et al.[57] achieved LOD values for PAH derivatives in sub-pg

1732https://www.wendangku.net/doc/7b15222231.html,/locate/trac

(oxy-PAHs)to sub-ng range (nitro-PAHs)in both parti-cle and gas phase through application of a solvent-based GC-MS method.

Although the absolute LOD re?ects the most impor-tant feature of any analytical method,it can be manip-ulated to a certain degree by altering the initial volume of air sampled.Hence,for comparison of LODs in a practical sense,they should be compared in terms of both absolute mass and mass concentration (e.g.,ng/m 3).If the available research is surveyed from this aspect,most studies were able to quantify PAHs at ng/m 3or even pg/m 3levels (Tables 6and 7).

As another important QA parameter,reproducibility expressed in terms of relative standard deviation (RSD%)can be examined across different studies.If the data on PAHs are compared from this aspect,they vary greatly across individual compounds –as low as 2%in case of FLT and as high as 28%for ANT in a TD-GC-MS analysis of particle-phase PAHs in NIST SRM1649a (urban dust)[61].Likewise,it was below 10%for all the 16priority PAHs,except for PHN (20%)in both particle and gas phases,when analyzed by a solvent-based GC-MS method for indoor-air samples [40].Interestingly,in vapor phase,the lowest values were 3%(PHN),while the highest was 14%(PYR)[18];these authors used a microwave-assisted HS-SPME-GC-MS method.

Wauters et al.[49]assessed the reproducibility of their TD-GC-MS analysis for 16priority PAHs and found them in the range 3.7%(PYR)to 12.9%(NAP),with an average of 7.4%.Hence,the reproducibility of PAH analysis is generally seen to be maintained near or below the 10%level (RSD).However,on certain occasions,this TD-GC-MS method suffered from relatively poor repro-ducibility,especially for lighter PAHs,regardless of phase type.

Although still fairly rare,some studies aimed to com-pare the relative performance between emerging TD-based methods and conventional solvent-based methods as a means to warrant reliability in PAH determination.Wauters et al.[49]compared their TD-GC-MS method to the classical method for quantitative analysis of the 16priority PAHs.They found that concentrations measured using the new method were signi?cantly higher than those obtained using the classical method (i.e.a factor 1.2–3for the high molecular weight PAHs).Moreover,those differences were extended further for lighter PAHs up to 35times for NAP and 23times for ACL.

In another study,Van Drooge et al.[59]assessed the detectability of both conventional solvent extraction and a new TD-based method interfaced with a GC-MS system.The TD method was reported to have lower uncertainties than the conventional solvent-based extraction method,if the analysis of more reactive PAHs is made at similar concentrations (e.g.,BAP);the use of TD methodology instead of the solvent-based method enabled the reduc-tion of the analytical expanded uncertainty from 19%to

T a b l e 6.(c o n t i n u e d )

T a r g e t P A H s m o d e a

E x t r a c t i o n (S o l v e n t -b a s e d o r S o l v e n t f r e e )

P r e t r e a t m e n t a n d s a m p l e l o a d i n g

D e t e c t i o n m e t h o d L O D

P r e c i s i o n

R e c o v e r y

R e f .

F l o w r a t e (L /m i n 1)

S a m p l i n g v o l u m e (L )A b s o l u t e m a s s (n g )

C o n c e n t r a t i o n (n g /m 3)

(%R S D )

(%)

D

S o x h l e t e x t r a c t i o n w i t h h e x a n e D I

G C -M S

450,000–900,000[31]

S o x h l e t e x t r a c t i o n w i t h h e x a n e

D I G C -M S 600

900,000

0.001

[32]

E

S o x h l e t e x t r a c t i o n w i t h d i c h l o r o m e t h a n e D I

G C -M S 900,000–1000,000[35]

P r e s s u r i z e d l i q u i d e x t r a c t i o n w i t h d i c h l o r o m e t h a n e /L i q u i d c h r o m a t o g r a p h y p u r i ?c a t i o n w i t h S P E c a r t r i d g e s

C o o l s p l i t l e s s i n j e c t i o n (40–320°C )G C -M S 1250

0.01-2.60p g (o x P A H s )a n d 0.03-0.07(n i t r o P A H s )

14–84(N i t r o P A H s )&5-83(o x P A H s )

[57]

a

T a r g e t m o d e (A =s o m e o f 16p r i o r i t y P A H s d e s i g n a t e d b y U S E P A ,B =s o m e o f 16+s o m e e x t r a ,C =16P A H s o n l y ,D =16P A H s +s o m e e x t r a ,a n d E =A l l t h e P A H s e x c e p t 16P A H s ).

Trends Trends in Analytical Chemistry,Vol.30,No.11,2011

1734

https://www.wendangku.net/doc/7b15222231.html,/locate/trac

红外热像检测技术综述

作业一红外热像检测技术综述 院（系）名称机械工程及自动化学院科目现代无损检测技术 学生姓名X X 学号XXXXXXXX 2016 年1X 月1X 日

红外热像检测技术综述 XXXX XXXX 目录 1 红外热像检测技术的原理介绍 (1) 2 红外热像检测技术的应用 (2) 2.1材料的内部制造缺陷的红外热像检测 (2) 2.3结构内部损伤及材料强度的检测 (3) 2.4在建筑节能检测中的应用 (3) 2.5建筑外外墙面饰面层粘贴质的检测 (4) 2.6在建筑物渗漏检测中的应用[13] (4) 3 红外热像检测技术国内外发展现状 (5) 3.1红外热像检测技术国外发展现状 (5) 3.2红外热像检测技术国内发展现状 (7) 4 参考文献 (10) I

1 红外热像检测技术的原理介绍 红外热成像检测技术采用主动式控制加热激发被检物内部缺陷，通过快速热图像采集和基于热波理论图像处理技术实现缺陷检测。它通过光学机械扫描系统，将物体发出的红外线辐射汇聚在红外探测器上，形成红外热图像，由此来分辨被测物体的表面温度。该技术具有检测速度快、非接触、范围广、精度高、易于实现自动化和实时观测等诸多优点，适合于裂缝、分层、积水、冲击损伤等问题的诊断。 红外线和可见光及无线电波一样是一种电磁波，红外线的波长比可见光长，比无线波短，为0.78~1000m μ，可分为近红外、中红外和远红外。任何物体只要不是绝对零度，都会因为分子的东{转和振动而发出“辐射能量”，红外辐射是其中一种。如果把物体看成是黑体，吸收所有的人射能量，则根据斯蒂芬—玻尔兹曼定律，在全波长范围内积分可得到黑体的总辐射度为: ()40 ,M M T d T λλσ∞==? (1.1) 式中:()()152121,exp 1c M T c W m m T λλμλ---??????=-???? ?????? ??? 为黑体的光谱辐射度;1c ，2c 为辐射常数，8241 3.741810c W m m μ-=???，42=1.438810c m K μ??，σ为斯蒂芬—玻尔兹曼常数，8245.6710W m K σ---=???，实际的大部分人工或天然材料都是灰体而不是黑体材料，与黑体不同，灰体材料的发射率1ε≠，灰体表面能反射一部分入射的长波()>3m λμ辐射，因此灰体表面的辐射由自身发射的和环境反射的两部分组成，用红外探测器可直接测量灰体发射和反射的总和ap M ，但无法确定各自的份额。通常假设物体表面为黑体，将ap M 称为表观辐 射度，为便于理解，一般将其转换为人们较熟悉的温度单位，称为表观温度ap T ，即: ()()()()04,,ap t l ap ap M M T M T d T λελλρλλλσ=+=? (1.2) 上述的表观温度ap T ，即为红外探测器测量所得温度。在无损检测中测量距离一般较近，可以忽瞬大气的影响，故被测物体的表面发射率。的取值是否准确是影响测量精度的关键因素。

微生物检测技术在食品检测中的应用研究进展 文献综述

微生物检测技术在食品检测中的应用研究进展 摘要：食品问题关系国计民生，食品的安全越来越受到人们关注。在食品工业迅速发展的今天，建立食品微生物快速检测方法，对食品质量进行检测、监控尤为重要。近几年各国的许多机构和学者都很重视食品微生物检测技术和方法的研究，本文对此进行了详细的介绍。 关键词：检测技术微生物食品安全 Progress of the research on the application of Microbial Detection Technology in food testing Abstract:Food is the people's livelihood. Food safety has received more and more attention. At present, the food industry is developing rapidly. Therefore, developing an rapid testing method of Microorganism in the food is especially important in detection and monitoring of food quality. In recente years, many institutes and researchers from different country attach great importance to the research of food microbiological testing techniques and methods. This article will give a detailed introduction to this below. Key words:the testing techniques Microorganism food safety 1前言 随着时代的不断发展，人们生活水平不断提高，食品安全问题也越来越受到人们的关注，近几年来，三聚氰氨、苏丹红、漂白剂等等一系列的食品安全问题使人们对食品产生了强烈的不信任感，因此，食品微生物检测技术的应用也越来越广泛，同时，食源性微生物的检测技术也趋向迅捷、准确、大通量的方向发展。以往的食品微生物检测技术已经无法应对现代的食品安全问题，检测速度缓慢、检测精度不精确，因此，应当采取新的食品微生物检测技术，现代的检测技术包括色谱法与荧光分析法、阻抗法、放射测量法、ELISA法和生物传感器法，结合我国实际情况，在建立标准的食源性微生物检测方法，推广标准化、检测技术的应用等方面还要很多工作要做[1]。 2 食品微生物检验的内容和特点 2.1 食品的污染程度指示菌的检验 （1）细菌总数：又称菌落数，是判断食物和应用水污染的主要指标。这是一种可以为卫生学检验评价提供依据的方法。 （2）大肠杆菌：这种细菌主要是来自人们本身的粪便，所以对大肠杆菌的数量来检验食物或饮

碰撞检测

二维碰撞检测算法 碰撞检测（Collision Detection,CD）也称为干涉检测或者接触检测，用来检测不同对象之间是否发生了碰撞，它是计算机动画、系统仿真、计算机图形学、计算几何、机器人学、CAD\ CAM等研究领域的经典问题。 碰撞物体可以分为两类：面模型和体模型。面模型是采用边界来表示物体，而体模型则是使用体元表示物体。面模型又可根据碰撞后物体是否发生形变分为刚体和软体，刚体本身又可根据生成方式的不同分为曲面模型和非曲面模型。目前对于碰撞的研究多集中于面模型的研究，因为体模型是一种三维描述方式，对它进行碰撞检测代价较高。而在面模型的研究中，对刚体的研究技术更为成熟。 下面列举几种常用的碰撞检测技术： 1：包围盒（bounding box）是由Clark提出的，基本思想是使用简单的几何形体包围虚拟场景中复杂的几何物体，当对两个物体进行碰撞检测时，首先检查两个物体最外层的包围盒是否相交，若不相交，则说明两个物体没有发生碰撞，否则再对两个物体进行检测。基于这个原理，包围盒适合对远距离物体的碰撞检测，若距离很近，其物体之间的包围盒很容易相交，会产生大量的二次检测，这样就增大了计算量。 包围盒的类型主要有AABB(Aligned Axis Bounding Box)沿坐标轴的包围盒、包围球、OBB(Oriented Bounding Box)方向包围盒和k-DOP(k Discrete Orientation Polytopes)离散方向多面体等。 AABB是包含几何对象且各边平行于坐标轴的最小六面体，两个AABB包围盒相交当且仅当它们三个坐标轴上的投影均重叠，只要存在一个方向上的投影不重叠，那么它们就不相交。AABB间的相交测试和包围体的更新速度比其他算法

基于单目视觉的路面车辆检测及跟踪方法综述

第24卷 第12期2007年12月 公 路 交 通 科 技 Journal of Highway and Transportation Research and Development Vol 24 No 12 Dec 2007 文章编号:1002 0268(2007)12 0127 05 收稿日期:2006 08 01 基金项目:江苏省科技计划高技术研究项目(BG2005008) 作者简介:胡铟(1973-),男,江西南昌人,博士研究生,研究方向为计算机视觉、目标检测及跟踪 (huyinyx@163 com) 基于单目视觉的路面车辆 检测及跟踪方法综述 胡 铟,杨静宇 (南京理工大学,江苏 南京 210094) 摘要:首先介绍了车辆检测算法的3种基本组成部分:检测、验证、跟踪,然后根据算法的组成重点介绍了车辆检测以及跟踪的几种主要算法。车辆检测算法包括基于特征的方法、基于光流场的方法和基于模型的方法,车辆跟踪算法包括基于区域相关的方法、基于活动轮廓的方法、基于特征的方法和MeanShift 快速跟踪算法。根据试验结果对各种车辆检测和跟踪方法的优点、缺点以及实际应用中不同情况下适用范围的局限性进行了综合分析。最后在结论部分总结展望了文中介绍的几种车辆检测和跟踪方法的应用前景,并提出了在实际应用时的一些建议和将来的主要研究和发展方向。 关键词:智能运输系统;车辆检测;单目视觉;跟踪中图分类号:TP391 4 文献标识码:A Veh icle D etection and Tracking Based on Monocu lar Vision HU Yin,YANG Jing yu (Nanjing Universi ty of Science &Technology,Jiangsu Nanjing 210094,China) Abstract :First,the three component of the vehicle detection algori thm including detection, verification and tracking are discussed Then,the algorithm of detection and tracking are discussed with emphasis on composition The vehicle detection algorithm includes feature based,op tical flow based and model based method The vehicle tracking al gorithm includes region correlation based,active contours based,feature based and mean shift based method The meri t and di sadvantage of these algori th ms is discussed accordin g to the result of experimentation Finally,some suggestions for fu ture research and application are presented Key words :Intelligent Transport Systems;vehicle detection;monocular visi on;trackin g 0 引言 近年来随着计算机视觉技术的发展,计算机视觉 在智能交通系统中得到了广泛的应用,如交通事件及流量的监测 [1] 、路面病害检测以及智能车辆的自动导 航等。作为智能交通系统的一个方面,智能车辆利用检测和智能算法去理解车辆的即时环境,并且提示驾驶员部分或完全控制车辆的行驶。 智能车辆的应用领域可以分为: (1)为驾驶员提供建议或警告(碰撞报警)。(2)部分的控制车辆,可以是持续的驾驶辅助, 如行道线的维持,或者是紧急事件的干预,碰撞的紧急避免措施。 (3)完全的控制车辆(自动驾驶系统)。 在过去的几年中,为了研究改良安全性和防止事故的新技术,许多国家和国际间的项目开始启动。车辆事故的统计数据揭示出其他车辆是驾驶员面临的主要威胁。因此研究对驾驶员发出关于行驶环境和可能与其他车辆碰撞的警告辅助系统受到更多的关注。 利用光学传感器的车辆检测是一个极富挑战性的任务,具体说有如下需要解决的难点问题: (1)车型多样:各种形状,大小,颜色;

图像分割方法综述

图像分割方法综述

图像分割方法综述 摘要：图像分割是计算计视觉研究中的经典难题，已成为图像理解领域关注的一个热点，本文对近年来图像分割方法的研究现状与新进展进行了系统的阐述。同时也对图像分割未来的发展趋势进行了展望。 关键词：图像分割；区域生长；活动边缘；聚类分析；遗传算法 Abstract:Image segmentation is a classic problem in computer vision,and become a hot topic in the field of image understanding. the research actuality and new progress about image segmentation in recent years are stated in this paper. And discussed the development trend about the image segmentation. Key words: image segmentation; regional growing; active contour; clustering

analysis genetic algorithm 1 引言 图像分割是图像分析的第一步，是计算机视觉的基础，是图像理解的重要组成部分，同时也是图像处理中最困难的问题之一。所谓图像分割是指根据灰度、彩色、空间纹理、几何形状等特征把图像划分成若干个互不相交的区域，使得这些特征在同一区域内表现出一致性或相似性，而在不同区域间表现出明显的不同。简单的说就是在一副图像中，把目标从背景中分离出来。对于灰度图像来说，区域内部的像素一般具有灰度相似性，而在区域的边界上一般具有灰度不连续性。 关于图像分割技术，由于问题本身的重要性和困难性，从20世纪70年代起图像分割问题就吸引了很多研究人员为之付出了巨大的努力。虽然到目前为止，还不存在一个通用的完美的图像分割的方法，但是对于图像分割的一般性规律则基本上已经达成的共识，已经产生了相当多的研究成果和方法。本文根据图像发展的历程，从传统的图像分割方法、结合特定工具的图像分割方

入侵检测技术综述

入侵检测技术综述 胡征兵1Shirochin V.P.2 乌克兰国立科技大学 摘要 Internet蓬勃发展到今天，计算机系统已经从独立的主机发展到复杂、互连的开放式系统，这给人们在信息利用和资源共享上带来了很大的便利。由Internet来传递和处理各种生活信息，早已成为人们重要的沟通方式之一，随之而来的各种攻击事件与入侵手法更是层出不穷，引发了一系列安全问题。本文介绍现今热门的网络安全技术-入侵检测技术，本文先讲述入侵检测的概念、模型及分类，并分析了其检测方法和不足之处，最后说描述了它的发展趋势及主要的IDS公司和产品。 关键词入侵检测入侵检测系统网络安全防火墙 1 引言 随着个人、企业和政府机构日益依赖于Internet进行通讯，协作及销售。对安全解决方案的需求急剧增长。这些安全解决方案应该能够阻止入侵者同时又能保证客户及合作伙伴的安全访问。虽然防火墙及强大的身份验证能够保护系统不受未经授权访问的侵扰，但是它们对专业黑客或恶意的经授权用户却无能为力。企业经常在防火墙系统上投入大量的资金，在Internet入口处部署防火墙系统来保证安全，依赖防火墙建立网络的组织往往是“外紧内松”，无法阻止内部人员所做的攻击,对信息流的控制缺乏灵活性，从外面看似非常安全，但内部缺乏必要的安全措施。据统计，全球80%以上的入侵来自于内部。由于性能的限制，防火墙通常不能提供实时的入侵检测能力，对于企业内部人员所做的攻击，防火墙形同虚设。 入侵检测是对防火墙及其有益的补充，入侵检测系统能使在入侵攻击对系统发生危害前，检测到入侵攻击，并利用报警与防护系统驱逐入侵攻击。在入侵攻击过程中，能减少入侵攻击所造成的损失。在被入侵攻击后，收集入侵攻击的相关信息，作为防范系统的知识，添加入知识库内，增强系统的防范能力，避免系统再次受到入侵。入侵检测被认为是防火墙之后的第二道安全闸门，在不影响网络性能的情况下能对网络进行监听，从而提供对内部攻击、外部攻击和误操作的实时保护,大大提高了网络的安全性[1]。 2 入侵检测的概念、模型 入侵检测（Intrusion Detection，ID）, 顾名思义，是对入侵行为的检测。它通过收集和分析计算机网络或计算机系统中若干关键点的信息，检查网络或系统中是否存在违反安全策略的行为和被攻击的迹象。进行入侵检测的软件与硬件的组合便是入侵检测系统（Intrusion Detection System,IDS）。 入侵检测的研究最早可以追溯到詹姆斯·安德森[1]在1980年为美国空军做的题为《计算机安全威胁监控与监视》的技术报告，第一次详细阐述了入侵检测的概念。他提出了一种对计算机系统风险和威胁的分类方法，并将威胁分为外部渗透、内部渗透和不法行为三种，还提出了利用审计跟踪数据监视入侵活动的思想。他的理论成为入侵检测系统设计及开发的基础 , 他的工作成为基于主机的入侵检测系统和其它入侵检测系统的出发点。 Denning[2]在1987年所发表的论文中，首先对入侵检测系统模式做出定义：一般而言，入侵检测通过网络封包或信息的收集，检测可能的入侵行为，并且能在入侵行为造成危害前及时发出报警通知系统管理员并进行相关的处理措施。为了达成这个目的，入侵检测系统应包含3个必要功能的组件：信息来源、分析引擎和响应组件。 ●信息来源（Information Source）：为检测可能的恶意攻击，IDS所检测的网络或系统必须能提供足够的信息给IDS，资料来源收集模组的任务就是要收集这些信息作为IDS分析引擎的资料输入。 ●分析引擎（Analysis Engine）：利用统计或规则的方式找出可能的入侵行为并将事件提供给响应组件。 ●响应模组（Response Component）：能够根据分析引擎的输出来采取应有的行动。通常具有自动化机制，如主动通知系统管理员、中断入侵者的连接和收集入侵信息等。 3 入侵检测系统的分类 入侵检测系统依照信息来源收集方式的不同，可以分为基于主机（Host-Based IDS）的和基于网络（Network-Based IDS）；另外按其分析方法可分为异常检测（Anomaly Detection,AD）和误用检测（Misuse Detection,MD），其分类架构如图1所示： 图 1. 入侵检测系统分类架构图

目标跟踪相关研究综述

Artificial Intelligence and Robotics Research 人工智能与机器人研究, 2015, 4(3), 17-22 Published Online August 2015 in Hans. https://www.wendangku.net/doc/7b15222231.html,/journal/airr https://www.wendangku.net/doc/7b15222231.html,/10.12677/airr.2015.43003 A Survey on Object Tracking Jialong Xu Aviation Military Affairs Deputy Office of PLA Navy in Nanjing Zone, Nanjing Jiangsu Email: pugongying_0532@https://www.wendangku.net/doc/7b15222231.html, Received: Aug. 1st, 2015; accepted: Aug. 17th, 2015; published: Aug. 20th, 2015 Copyright ? 2015 by author and Hans Publishers Inc. This work is licensed under the Creative Commons Attribution International License (CC BY). https://www.wendangku.net/doc/7b15222231.html,/licenses/by/4.0/ Abstract Object tracking is a process to locate an interested object in a series of image, so as to reconstruct the moving object’s track. This paper presents a summary of related works and analyzes the cha-racteristics of the algorithm. At last, some future directions are suggested. Keywords Object Tracking, Track Alignment, Object Detection 目标跟踪相关研究综述 徐佳龙 海军驻南京地区航空军事代表室，江苏南京 Email: pugongying_0532@https://www.wendangku.net/doc/7b15222231.html, 收稿日期：2015年8月1日；录用日期：2015年8月17日；发布日期：2015年8月20日 摘要 目标跟踪就是在视频序列的每幅图像中找到所感兴趣的运动目标的位置，建立起运动目标在各幅图像中的联系。本文分类总结了目标跟踪的相关工作，并进行了分析和展望。

自动检测技术综述

课程综述 08自动化（2）班 0805070124 随着大三分专业被分到自动化这个专业后，首先接触的课程就是自动化检测技术。作为一门动手能力要求比较高的课程，这门课也配套了相应的实验课程。回味这学期的自动检测技术这门课的学习，按照自己的学习计划学习起来还是比较从容的。自动检测技术是自动化科学技术的一个重要分支科学，是在仪器仪表的使用、研制、生产、的基础上发展起来的一门综合性技术。自动检测就是在测量和检验过程中完全不需要或仅需要很少的人工干预而自动进行并完成的。实现自动检测可以提高自动化水平和程度，减少人为干扰因素和人为差错，可以提高生产过程或设备的可靠性及运行效率。 对于自动检测技术这门课，我觉得很重要的就是学习计划，不管做什么事都应该有一个计划，大到自己的学习生涯规划，小到自己的一天什么时刻该做什么，这样你才能做到有的放矢。作为一门自动化专业重要的课程，学好这门课是必须的。关于学习计划，我觉得每天课后的复习工作是很重要，这有助于我们对新知识的理解和吸收。大学里要充分利用各种资源，比如说图书馆、学术论坛、网络资源等。网络这种全新的学习形式具有开放性、互动性、网络性、虚拟性的特点,为我们的自主学习，教师的教学提供了许多便利条件。目前，互联网上学习资源中，管理方面的资源极为丰富；收费、互助、免费应有尽有。当然如何有效利用这些资源，是我们必须重视的问题，不适当的选择，会浪费精力，浪费时间，我们要选择适合自己的资源进行学习，这样才能做到事半功倍。还有就是一个老师每次布置的小设计论文，每次的设计论文感觉都是对自己所学到的知识的一些升华，在原有知识的基础上进行设计，利用课本知识缩学到的东西应用到实际的设计当中去。这是一项很好的作业，让我们在学到基础知识的前提下，能够活用。 自动检测的任务：自动检测的任务主要有两种，一是将被测参数直接测量并显示出来，以告诉人们或其他系统有关被测对象的变化情况，即通常而言的自动检测或自动测试；二是用作自动控制系统的前端系统，以便根据参数的变化情况做出相应的控制决策，实施自动控制。自动检测技术主要的研究内容：自动检测技术的主要研究内容包括测量原理、测量方法、测量系统、及数据处理。测量系统：确定了被测量的测量原理和测量方法后，就要设计或选用装置组成测量系统。目前的测量系统从信息的传输形式看，主要有模拟式和数字式两种。模拟式测量系统：模拟量测试系统是由传感器，信号调理器，显示、记录装置和（或）输出装置组成。数字式测量系统：数字式测量系统目前主要是带微机的测量系统，是由传感器、信号调理器、输入接口、中央处理器组件、输出接口和显示记录等外围设备组成。检测技术的特点：实时性强、精确度高、可靠性高、通道多、功能强。 关于自动化专业而言，自动化检测技术对学生动手能力要求比较高，所以有关和自动检测技术的实验课也是尤为重要的一项。实验课的课前准备工作必须充

机器视觉技术发展现状文献综述

机器视觉技术发展现状 人类认识外界信息的80%来自于视觉，而机器视觉就是用机器代替人眼来做 测量和判断，机器视觉的最终目标就是使计算机像人一样，通过视觉观察和理解 世界，具有自主适应环境的能力。作为一个新兴学科，同时也是一个交叉学科，取“信息”的人工智能系统，其特点是可提高生产的柔性和自动化程度。目前机器视觉技术已经在很多工业制造领域得到了应用，并逐渐进入我们的日常生活。 机器视觉是通过对相关的理论和技术进行研究，从而建立由图像或多维数据中获机器视觉简介 机器视觉就是用机器代替人眼来做测量和判断。机器视觉主要利用计算机来模拟人的视觉功能，再现于人类视觉有关的某些智能行为，从客观事物的图像中提取信息进行处理，并加以理解，最终用于实际检测和控制。机器视觉是一项综合技术，其包括数字处理、机械工程技术、控制、光源照明技术、光学成像、传感器技术、模拟与数字视频技术、计算机软硬件技术和人机接口技术等，这些技术相互协调才能构成一个完整的工业机器视觉系统[1]。 机器视觉强调实用性，要能适应工业现场恶劣的环境，并要有合理的性价比、通用的通讯接口、较高的容错能力和安全性、较强的通用性和可移植性。其更强调的是实时性，要求高速度和高精度，且具有非接触性、实时性、自动化和智能 高等优点，有着广泛的应用前景[1]。 一个典型的工业机器人视觉应用系统包括光源、光学成像系统、图像捕捉系统、图像采集与数字化模块、智能图像处理与决策模块以及控制执行模块。通过 CCD或CMOS摄像机将被测目标转换为图像信号，然后通过A/D转换成数字信号传送给专用的图像处理系统，并根据像素分布、亮度和颜色等信息，将其转换成数字化信息。图像系统对这些信号进行各种运算来抽取目标的特征，如面积、 数量、位置和长度等，进而根据判别的结果来控制现场的设备动作[1]。 机器视觉一般都包括下面四个过程：

虚拟手术中实时碰撞检测技术

虚拟手术中实时碰撞检测技术研究 彭 磊 张裕飞 王秀娟 （泰山医学院 信息工程学院 山东 泰安 271016） 摘 要： 碰撞检测是虚拟手术的关键技术，为提高检测速度，满足系统实时性的要求，提出空间剖分和层次包围盒相结合的方法。使用八叉树表示法对虚拟场景进行空间剖分，在叶节点构建层次包围盒。进行碰撞检测时属于不同八叉树节点的几何元素不会相交，否则使用层次包围盒算法继续进行检测，对于有可能相交的几何元素再进行精确相交检测。 关键词： 虚拟手术；碰撞检测；空间剖分；层次包围盒 中图分类号：TP391.9 文献标识码：A 文章编号：1671－7597（2012）1120029－02 进行碰撞检测时从八叉树的根节点开始，计算两几何元素0 引言 是否属于同一节点，如果不属于同一节点则不相交，如果属于虚拟手术是集医学、生物力学、材料学、计算机图形学、同一节点，递归的到下一级节点进行检查，直到发现两几何元虚拟现实等诸多学科为一体的交叉研究领域。虚拟手术在医学素属于同一叶节点，则需要进一步使用层次包围盒进行检查。 中的应用主要包括：手术计划与过程模拟、术中导航与监护、 2 层次包围盒 手术教学与训练等。碰撞检测是虚拟手术系统中的关键技术，贯穿于虚拟手术的整个过程。 对于八叉树的每个叶节点包含的几何元素，建立层次包围虚拟手术系统中的对象根据材质可分为刚体组织和软件组盒（Bounding Volume Hierarchy ，BVH ）。相对于单纯的层次织。骨骼、手术器械等属于刚体组织，而人体的许多器官如肌包围盒技术，使用空间剖分与层次包围盒相结合的方法进行碰肉、血管、肝脏等属于软体组织。以往大部分碰撞检测的研究撞检测，构建的层次树规模更小，计算量更少。层次包围工作都是针对刚体对象的。与刚体相比较，软体组织由于其特殊的物理性质，在外力或某些操作的作用下会发生几何形状、位置甚至数量上的变化，因此基于软体组织的碰撞检测需要更详细的信息和更多的处理。 最简单的碰撞检测方法是对场景中的几何元素进行两两相2交测试，其时间复杂度为O(n )，虽然这种方法可以得到正确的结果，但是当场景中的几何模型稍微增多些，其实时性便无法满足实际的需求。为了尽可能地减少参与相交测试的几何元素的数量，提高系统的实时性，目前碰撞检测技术使用的主要算法有：层次包围盒法，空间分割法，基于网格剖分的方法[1]。但是这些经典的算法也都存在着构造难度大、紧密性差、相交测试复杂、效率低等缺点。 本文采用空间剖分和层次包围盒相结合的方法，简化了几何信息的表示，进行碰撞检测时可排除明显不相交的几何元素，无法排除的再进行精确相交检测，从而减少计算量，加速碰撞检测速度，提高系统实时性。 1 空间剖分技术 整个虚拟手术的场景空间递归的剖分成若干个网格单元，每一个几何元素都属于某个网格单元，处于同一网格单元内的几何元素才有相交的可能，不在同一网格单元的几何元素一定不会相交。采用八叉树的表示方法进行空间剖分。即包含整个场景的立方体作为八叉树的根节点，立方体的3条棱边分别与x ，y ，z 轴平行。递归的将立方体剖分为8个小块，如图1（a ）所示，生成8个子节点，直到达到指定的剖分层次为止，如图1（b ）所示，每个叶节点包含有限个几何元素。 图1 八叉树表示法 盒包括包围盒和层次树两种数据结构。 2.1 包围盒 包围盒技术是减少相交检测次数，降低碰撞检测复杂度的一种有效的方法。其基本思想是用几何形状相对简单的封闭表面将一复杂几何元素包裹起来，首先进行包围盒之间的相交测试，排除明显不相交的几何元素，无法排除的几何元素，再进一步进行精确的相交测试，从而达到减少相交测试计算量的目的。常见的包围盒类型有：包围球（Bounding Sphere ）、沿坐标轴的包围盒（Axis Aligned Bounding Box ，AABB ）、方向包围盒（Oriented Bounding Box ，OBB ）。离散方向包围盒（k-Discrete Orientation Polytopes ，k-DOPs ）等[2]，如图2所示。 图2 包围盒 由于虚拟手术对实时性要求较高，本文选择AABB 型包围盒，AABB 是平行于坐标轴的，包含几何元素的最小正立方体。其优点是：1）易于构建，只需要计算所包含几何元素的顶点的x ，y ，z 坐标的最大值和最小值，存储6个浮点数即可；2）相交测试计算量小，相交测试时只需对两个包围盒在三个坐标轴上的投影分别进行比较，最多6次比较运算即可。 2.2 包围盒层次树 包围盒层次树即包围盒的层次结构，层次树的根节点包含某个八叉树叶节点几何元素的全集，向下逐层分裂，直到每个叶节点表示一个基本几何元素。常用的构建策略有自顶向下和自底向上两种。 自顶向下的方法首先建立根结点，利用基于全集的信息递归地将每个节点分裂为两个或多个子集，直至生成只包含一个 基本图元的叶结点为止，从而建立一棵自顶向下的包围盒层次 （ ）八叉树结构 （ ）节点的剖分

文献综述

文献综述 前言 近年来,除测绘领域之外,其他行业如机械制造、建筑、医学等,对近景摄影测量技术的需求也越来越大,传统的量测相机显然已经无法满足其要求,而越来越普及的非量测相机正好可以填补这个空缺,利用非量测相机进行摄影测量具有非常远大的应用前景。非量测相机在影像获取方面具有使用简单方便、价格合理，作业效率高、适应性强等优点，但是非量测相机的主距f和像主点在像平面坐标系中的坐标(x0,y0)都是未知的,并且非量测相机存在较大的镜头畸变,因此必须先对其进行检校,然后才能进行后续的像点量测和数据处理。 所谓的相机检校是指借助于像平面上一些点在物方坐标系中的坐标，确定照相机的内、外参数，得到有效的成像模型，以达到在像平面上像素点与三维空间中的点之间建立映射的目的。广义上讲，近景摄影机检校的内容包括： 1.主点位置(x0,y0)与主距(f)的测定; 2.光学畸变系数的测定; 3.压平装置以及像框坐标系的设定; 4.调焦后主距变化的测定与设定; 5.调焦后畸变差变化的测定; 6.摄影机偏心常数的测定; 7.立体摄影机(及立体视觉系统)内方位元素与外方位元素的测定;

8.多台摄影机同步精度的测定。 对于一般相机检校任务，我们主要测定相机的内方位元素（x0,y0,f）和镜头畸变差参数（k1,k2,p1,p2），其主要的检校方法大体可以分为：光学实验室检校(Optical Laboratory Calibration)法；实验场检校(Test Range Calibration)法；作业检校(On the Job Calibration)法；自检校(Self Calibration)法；恒星检校(Stellar Calibration)法。 其中，适用于非量测相机检校的作业检校法是一种在完成某个近景测量任务中同时对相机进行检校的方法。此方法依据物方空间分布合理的一群高质量控制点,在解求待定点物方空间坐标的任务中,同时解求像片内外方位元素、物镜畸变系数。基于直接线性变换（DLT）的相机检校法,也属作业检校法的一种。 直接线性变换(direct linear t ransformation ,简称DL T) 是建立像点坐标和物点坐标直接线性关系的算法。处理时不需要相机内外方位元素的初始值,因而在近景摄影测量中被广泛应用。目前市场上，利用DLT方法进行相机检校的算法,多是在有3维控制信息的情况下,采 用3维DLT方法进行检校较多,而采用2维DLT方法的较少。利用3维控制场进行相机检校,程序虽简单,但高精度3维控制场的建立比较困难。廉价数码相机的广泛应用,促使简便、高精度的检校方法成为近年来研究的主要方向。随着摄影测量和计算机视觉理论的发展,许多学者对相机检校技术进行了深入的研究,检校用的控制场也由3维向2维转变。基于2维DLT相机标定算法的研究已成为近几年相机标定研究的热点。

图像分割方法综述

图像分割方法综述 摘要：图像分割是计算计视觉研究中的经典难题，已成为图像理解领域关注的一个热点， 本文对近年来图像分割方法的研究现状与新进展进行了系统的阐述。同时也对图像分割未来的发展趋势进行了展望。 关键词：图像分割；区域生长；活动边缘；聚类分析；遗传算法 Abstract: Image segmentation is a classic problem in computer vision,and become a hot topic in the field of image understanding. the research actuality and new progress about image segmentation in recent years are stated in this paper. And discussed the development trend about the image segmentation. Key words: image segmentation; regional growing; active contour; clustering analysis genetic algorithm 1 引言 图像分割是图像分析的第一步，是计算机视觉的基础，是图像理解的重要组成部分，同时也是图像处理中最困难的问题之一。所谓图像分割是指根据灰度、彩色、空间纹理、几何形状等特征把图像划分成若干个互不相交的区域，使得这些特征在同一区域内表现出一致性或相似性，而在不同区域间表现出明显的不同。简单的说就是在一副图像中，把目标从背景中分离出来。对于灰度图像来说，区域内部的像素一般具有灰度相似性，而在区域的边界上一般具有灰度不连续性。 关于图像分割技术，由于问题本身的重要性和困难性，从20世纪70年代起图像分割问题就吸引了很多研究人员为之付出了巨大的努力。虽然到目前为止，还不存在一个通用的完美的图像分割的方法，但是对于图像分割的一般性规律则基本上已经达成的共识，已经产生了相当多的研究成果和方法。本文根据图像发展的历程，从传统的图像分割方法、结合特定工具的图像分割方法、基于人工智能的图像分割方法三个由低到高的阶段对图像分割进行全面的论述。 2 传统的图像分割方法 2.1 基于阀值的图像分割方法 阀值分割法是一种传统的图像分割方法，因其实现简单、计算量小、性能较稳定而成为图像分割中最基本和应用最广泛的分割技术。阀值分割法的基本原理是通过设定不同的特征阀值，把图像像素点分为具有不同灰度级的目标区域和背景区域的若干类。它特别适用于目标和背景占据不同灰度级范围的图，目前在图像处理领域被广泛应用，其中阀值的选取是图像阀值分割中的关键技术。 灰度阀值分割方法是一种最常用的并行区域技术，是图像分割中应用数量最多的一类。图像若只用目标和背景两大类，那么只需要选取一个阀值，此分割方法称为单阀值分割。单阀值分割实际上是输入图像f到输出图像g的如下变换：

既有建筑检测技术综述

既有建筑检测技术综述 摘要：建筑检测对建筑结构的安全性有着重要影响，因此在对建筑结构进行检测时必须要认真。建筑构件的检测工作与建筑的现在有着密切关系，因此相关部门需要对建筑工程的质量进行检测工作进行紧密监督，同时需要对建筑结构中的构建进行检测，本文在对既有建筑检测管理进行简单叙述的基础上，对既有建筑检测技术进行了详细分析，希望对相关工作人员能够有所帮助。 关键词：建筑检测;技术分析;建筑结构 经济的快速发展，使得我国的建筑行业得到了飞速发展，在建筑行业飞速发展的过程中，检测技术显得尤为重要。从目前情况来看，我国的建筑检测技术处于平稳发展阶段，前景一片大好。 一、建筑检测的管理 （一）单位资质 建筑检测单位必须具有相应的检测资质，只有具备检测资质的建筑单位才可以对建筑机构进行检测。检测单位需要应当为专门单位，其工作范畴只能是工程质量检测，不能建筑其它工作，确保建筑检测部门具有检测资质。 （二）检测人员的专业水平

从目前情况来看，我国的许多技术检测人员都通过了专业的技术培训，具有检测资格，但是仍然存在技术检测人员只是通过了简单的技术培训，并没有深入的了解建筑检测工作的评定标准和规范，因此在检测工作中难免会出现一些错误，这也导致了既有建筑检测技术无法在建筑检测中发挥应有的作用，因此必须要加强管理，通过技能培训提高建筑检测人员的专业素养，使既有检测检测技术在建筑检测中能够起到应有的作用。 （三）检测人员的职业素养 建筑检测人员的职业道德必须过关，建筑检测人员在进行建筑检测工作是，必须要公平、公正，不能盲目。但在实际工作中，有少数的检测人员在工作中为了实现某种目的或个人利益，人为的对检测结果进行修改，最终使检测结果出现偏差，这使得检测技术成为了口壳子，在建筑检测过程中并没有得到合理应用，影响了建筑检测结果。 二、既有建筑检测技术 （一）砌体检测技术 目前针对砂浆、砌体的检测方法较多，其中主要的检测方法有回弹法、贯入法、推出发等。但实际建筑中因为砂浆和砌体强度离散性大，因此在检测上的优劣势差异较为明显，因此在检测上经常会因为选用不同的检测方法而得到不同 的检测结构。下面重点分析贯入法，贯入法在检测过程中对

农药残留对食品安全的影响以及农药残留检测技术的文献综述

农药残留对食品安全的影响以及农药残留 检测技术的文献综述 摘要：介绍了农药残留的现状及其对食品安全的影响，同时对农药残留检测技术进行系统的综述，并对今后农药残留检测及控制进行了展望。 关键词：农药残留食品安全检测技术 农药残留是指在农业生产中施用农药后一定时期内残留于生物体、农副产品及环境中微量的农药原体、有毒代谢物、降解物和杂质的总称。残留的数量叫农药残留量，以每千克样本中有多少毫克（或微克、纳克等）表示。农药残留是使用农药后的必然现象，是不可避免的。农副产品上的残留量超过限量，人畜长期食用后会引起慢性中毒或病变，直接或间接影响人们的身体健康。因此，控制降低农药残留，发展农药残留检测技术已成为当前亟待解决的问题。 1农药残留现状及种类 1.1 农药残留的现状 “民以食为天，食以安为先”，农产品的质量安全直接关系到人们的健康和安全。在农业生产中，由于农药、化肥等农业化学投入品的使用，导致农作物严重污染，人们食用农药残留超标的农产品，引起食物中毒的事件经常发生。2010年1 月25 日至2 月5 日，武汉市农业局在抽检中发现来自海南省英洲镇和崖城镇的5个豇豆样品水胺硫磷农药残留超标，消息一出，立即引起社会各方关注，豇豆产地收购价与销售批发价均出现大幅下滑。 农药残留已经成为我国农产品出口的最大障碍，常常被进口国当作借口阻挡在门外，不仅给农户造成经济损失，而且还导致农产品出口竞争力减弱或下降，引起国家之间的经济贸易纠纷。国际市场对出口农产品安全要求很高，从2000 年起，欧盟等国家对农药残留颁布了更严格的标准，从2006 年5 月29 日开始，在日本市场流通的生鲜食品就适用肯定列表制度，一棵白菜要检测20个项目，最多的一种农产品要检测50个项目，合格后才能通关[1]。 农药喷洒在作物上经过一定时间后，由于日晒、雨淋、风吹、高温挥发和植物代谢等的作用，药剂逐渐分解、减少，但不能全部消失，收获的农副产品上仍

图像局部特征点检测算法综述

图像局部特征点检测算法综述 研究图像特征检测已经有一段时间了，图像特征检测的方法很多，又加上各种算法的变形，所以难以在短时间内全面的了解，只是对主流的特征检测算法的原理进行了学习。总体来说，图像特征可以包括颜色特征、纹理特等、形状特征以及局部特征点等。其中局部特点具有很好的稳定性，不容易受外界环境的干扰，本篇文章也是对这方面知识的一个总结。 本篇文章现在(2015/1/30)只是以初稿的形式，列出了主体的框架，后面还有许多地方需要增加与修改，例如2013年新出现的基于非线性尺度空间的KAZE特征提取方法以及它的改进AKATE等。在应用方面，后面会增一些具有实际代码的例子，尤其是基于特征点的搜索与运动目标跟踪方面。 1. 局部特征点 图像特征提取是图像分析与图像识别的前提，它是将高维的图像数据进行简化表达最有效的方式，从一幅图像的M×N×3的数据矩阵中，我们看不出任何信息，所以我们必须根据这些数据提取出图像中的关键信息，一些基本元件以及它们的关系。 局部特征点是图像特征的局部表达，它只能反正图像上具有的局部特殊性，所以它只适合于对图像进行匹配，检索等应用。对于图像理解则不太适合。而后者更关心一些全局特征，如颜色分布，纹理特征，主要物体的形状等。全局特征容易受到环境的干扰，光照，旋转，噪声等不利因素都会影响全局特征。相比而言，局部特征点，往往对应着图像中的一些线条交叉，明暗变化的结构中，受到的干扰也少。 而斑点与角点是两类局部特征点。斑点通常是指与周围有着颜色和灰度差别的区域，如草原上的一棵树或一栋房子。它是一个区域，所以它比角点的噪能力要强，稳定性要好。而角点则是图像中一边物体的拐角或者线条之间的交叉部分。 2. 斑点检测原理与举例 2.1 LoG与DoH 斑点检测的方法主要包括利用高斯拉普拉斯算子检测的方法（LOG）,以及利用像素点Hessian矩阵（二阶微分）及其行列式值的方法（DOH）。 LoG的方法已经在斑点检测这入篇文章里作了详细的描述。因为二维高斯函数的拉普拉斯核很像一个斑点，所以可以利用卷积来求出图像中的斑点状的结构。 DoH方法就是利用图像点二阶微分Hessian矩阵：