当前位置：文档库 › 环境空气污染物测量的溯源性

环境空气污染物测量的溯源性

Traceability of pollutant measurements for ambient air monitoring

V.Desauziers

Air-quality management is implemented by national monitoring networks. This implies quality assurance of analysis to ensure good traceability of the data obtained that is of great importance for data comparison on temporal and spatial scales and at national and international levels.In this aim,several tools have been established through regulation:de?nition of the priority pollutants to be monitored;determination of limit values;and, documented standards.These di?erent points are presented in this article and the particular case of measurement of volatile organic compounds (VOCs)in ambient air highlighted,through discrepancies observed between reference methods,the further e?orts that need to be made to achieve a better understanding of the variability of results and the need for developments to enhance air monitoring.

#2003Published by Elsevier B.V.

Keywords:Air monitoring;Analysis;Pollutants;Sampling;Traceability

1.Introduction

In recent decades,air quality has become

a very important concern as more and

more studies have shown the great

impact of atmospheric pollution on envir-

onment and health.

With di?erent processes,such as pre-

cipitation and in￠ltration,pollutants can

deposit onto soils or into natural waters,

reach groundwater and,hence,damage

ecosystems.We can highlight the main

phenomena of atmospheric pollution as

acidi￠cation,photo-oxidation,the green-

house e?ect and the impoverishment of

stratospheric ozone.

Acidi￠cation,through acid rains

(HNO3and H2SO4)are formed from nitro-

gen oxides and SO2,has been shown to

have an impact on forest and natural

water ecosystems and on building

erosion.Nitrogen oxides(NOx)are of

particular concern,as are volatile

organic compounds(VOCs),precursors of photochemical pollutants,such as ozone, aldehydes and peroxyl-acetyl-nitrate (PAN).

Compared with tropospheric ozone, which directly impacts human health, stratospheric ozone is necessary as UV ￠lter,and emissions of its main destruc-tors,chloro-￡uoro-carbons(CFCs), should be reduced.Depending on their concentrations,some of these pollutants may cause toxic e?ects to human beings, from nausea and breathing di?culties to cancer[1].This is particularly demon-strated in urban areas where most of the population of industrialized countries is concentrated(e.g.,in Europe,278mil-lion people live in cities with more than 50,000inhabitants,while378million live in rural areas[2]).

To improve air quality,regulation has been implemented on an international scale.It generally implies emission inven-tories,air monitoring and atmospheric dispersion modeling.To estimate emis-sions and elaborate management strate-gies,several methodologies have been developed.In Europe,the CORINAIR (CORe INventory of AIR emissions in Europe)helps quantify annual national releases into the atmosphere[3].A t the international level,because of the di?-culties of data comparison,an attempt to harmonize the methodologies has been going on between various international bodies(e.g.,European Commission(EC), the Intergovernmental Panel on Climate Change(IPCC),EUROSTAT,and the International Atomic Energy Agency) [3].

The traditional approach to evaluating air quality involves concentration

V.Desauziers*

Laboratoire Ge nie de

l’Environnement Industriel,

Ecole des Mines d’Ale s,

He lioparc,2avenue Pierre

Angot,F-64053Pau Cedex,

France

*Tel.:+33(0)5.59.30.54.25;

Fax:+33(0)5.59.30.63.68;

E-mail:Valerie.desauziers@

ema.fr

2520165-9936/03/$-see front matter#2003Published by Elsevier B.V.doi:10.1016/S0165-9936(04)00310-3

measurements in air and comparison with limit values. This is performed through monitoring sites(4300in North America)which belong to national networks (e.g.,State and Local Air Monitoring Stations(SLAMS) in the USA,and ATMO in France).Monitoring air quality through measurement networks implies quality assurance(QA)of analysis to ensure good traceability of the data obtained.Traceability is particularly impor-tant in this case because of the necessity for data com-parison on temporal and spatial scales,at national and international levels.In this aim,several tools have been established through regulation:

de￠nition of the priority pollutants to be monitored,in agreement with the ability of the

techniques to measure them;

determination of limit values that will condition the analytical steps and the method performance

(detection limits,repeatability...);

establishment of documented standards that describe analytical procedures with minimum

quality requirements;and,calibration proce-

dures.

These di?erent points will be discussed in the follow-ing sections.

Recently,in addition to the classical approach of data comparison with limit values described above, modeling has been included.New regulations,such as the EC’s Framework Directive(96/62/CE)allows the use of atmospheric dispersion modeling systems.They involve spatial and temporal scales,are applied to assess the future air quality against the air-quality objective[4]and can be consistent tools for air-quality management in case of poor spatial covering of mon-itoring data[5].However,these models should be care-fully validated and scienti￠c uncertainties should be assessed so as to obtain good agreement with the air-management policy[4].

2.De￠nition of pollutants to be monitored Priority pollutants were selected according to criteria that included both their known deleterious e?ects on health and the feasibility of their measurement,which is of the utmost importance for traceability.

For example,in Europe,air quality has been regu-lated since the early1980through EC Directives con-cerning SO2and Suspended Particle Matter(80/779/ EEC),lead(82/884/EEC),NO2(85/203/EEC)[2].In 1992,ozone was added(92/72/EEC).After15years of air quality measurements,it was demonstrated that these Directives were not as e?ective as expected.As a result,the EC published the new Framework Directive on Ambient Air Quality Assessment and Management (96/62/CE).The monitoring domain was enlarged to 13pollution indicators.They include pollutants already controlled by the Directives cited above and new compounds:benzene;polyaromatic hydrocarbons (PA Hs);carbon monoxide;cadmium;arsenic;and, mercury[2].One of the objectives of this new Directive is the uniform evaluation of air quality,so it involves improving traceability.To achieve this aim,for each pollutant a set of new Directives(‘‘Daughter Direc-tives’’)are being prepared through working groups that have responsibility for collect information about pollutant levels,health e?ects,and measurement methods and location.

In North America,the same pollutants are measured through4300monitoring sites[6].Among them,parti-culate matter generally corresponds to PM10,particles 10 m in diameter or smaller.In1997,the US Environ-mental Protection Agency(EPA)announced a new PM2.5standard as being necessary for health and environ-mental protection[6].This new standard has initiated a debate in the scienti￠c community regarding the appro-priate measurement to be considered and new rules are in progress[6].Through the1990Clean Air Act Amendments and in order to improve monitoring of ozone and its precursors(mainly VOCs),the US EPA initiated the Photochemical Assessment Monitoring Stations(PAMS)program in1993,and routine VOC measurements have been applied[6].As a result, individual measurements and therefore separative techniques are required,but long term on-site monitor-ing imposes a lot of technical constraints in this case. The analytical challenge is particularly strong.Other factors to be considered in selecting or developing ana-lytical procedures are the concentration levels to be quanti￠ed and the frequency of measurements.These are described in the following section for each priority pollutant.

3.Limit values and concentration units

The new proposal in the EC Directive includes require-ments in terms of limit values and margin of tolerance. Both in Europe and North America,measurement frequencies and limit values have been de￠ned with respect to the impact of the pollutant on human health (1h,8h,24h and annual averaging periods)or on vege-tation(annual averaging period)(see Table1).

The limit values de￠ned by the Framework Directive should not be exceeded from2005for SO2,CO and PM10,and from2010for NO2and benzene.Until these dates,margins of tolerance are allowed:43%for SO2, 50%for NO2,PM10and CO,and100%for benzene. These margins will progressively reduce every 12months by equal annual percentages to reach0%by January2005or2010according to the pollutant.

https://www.wendangku.net/doc/b9861860.html,/locate/trac253

https://www.wendangku.net/doc/b9861860.html,ment on concentration units

Table1shows that,whether considering EC Directives or US NAAQS,the concentration units of pollutants in air are di?erent( g/m3or ppb(or ppm)).By contrast, for water pollutant the concentration units are the same( g/L and ppb,as water density is1).For air,ppb or ppm(also noted ppbv and ppmv)corresponds to the ratio‘‘pollutant volume/air volume’’,which is there-fore temperature-and pressure-dependent and is not equivalent to g/m3or mg/m3.For example,40 g/m3 NO2corresponds to44.6ppbv at1atm and273K.In this case,the result can be also expressed as40 g/Nm3 where‘‘N’’is for normal conditions of pressure and temperature.This di?erence in concentration units can cause mistakes and di?culties for data inter-compar-isons and hence can a?ect traceability.

4.Reference methods(RMs)

For most of the criteria pollutants listed in Table1,the limit values to be reached are low(ppb or g/m3levels). As a result,analytical instruments need to be sensitive enough and to be able to provide quantitative results within the recommended averaging times;that generally requires routine measurements on site.

To ensure traceability of the measurements per-formed through monitoring networks,standard operat-ing procedures(SOPs),equivalent measurement methods and QA/QC(quality control)procedures should be implemented.The typical instrumentation and/or measurement methods described in inter-national,European and US standards are summarized in Table2.

These RMs have been elaborated and improved in the past10years.Development of RMs for suspended parti-cles(0.010^20 m in diameter),heavy metals(Pb,Cd, As,Ni),PAHs and ozone precursors are in progress in Europe and measurement of PM2.5is under develop-ment in USA.

The methods cited in Table2operate under estab-lished QArequirements.For SO2,NO2and O3,these requirements consist of routine zero,span and preci-sion checks,and periodic audits.Considering individual identi￠cation by gas chromatography(GC)analysis(for benzene and VOCs determination),routine calibration with single or multi-component standards is required. Inter-comparison exercises are also organized to vali-date RMs(e.g.,the HAMAQ(Harmonization of Air Quality Measurements for Important Atmospheric Pol-lutants in Europe)program that was held from1996to 1999,has allowed comparison of the RMs of seven European laboratories for￠ve pollutants(SO2,NO, NO2,benzene,and CO)[7].

However,despite these procedures being imple-mented to reach the quality objectives de￠ned in regulations,some monitoring issues remain,as illustrated in the following examples.

4.1.Performance of commercial instrumentation Some standard commercial analyzers are not sensitive or speci￠c enough when they operate in routine monitoring.This is the case for the measurement of nitrogen oxides by chemiluminescence[6]

. 254https://www.wendangku.net/doc/b9861860.html,/locate/trac

The principle is based on a reaction between NO and ozone to form excited NO2,which relaxes to its funda-mental energetic level by emitting a speci￠c radiation in the near infrared(NIR)(1200nm).In order to measure NO2,NO2is￠rst reduced in NO with a catalytic con-verter.Both NO and NO2(NO x)are then measured and, by di?erence with NO,the NO2concentration can be obtained.

Two shortcomings a?ect the use of these analyzers [6]:the converter is not speci￠c(nitric acid and organic nitrates can react);and,overestimation of the NO2con-centration can occur.

Moreover,the detection limit for NO is not low enough to evaluate correctly the NO2/NO ratio during periods of high ozone production.Alternatives have been developed,but they should be validated in routine monitoring[6].

4.2.Representative sampling

Because of the complexity of the sample,no completely satisfactory way of measurement can be found,even if RMs exist.This is the case for the measurement of sus-pended particles.RMs with10%precision are well described[9,10]for the measurement of particulate matter remaining on a￠lter after sampling.However, these RMs are subject to positive and negative artifacts because of the complexity of the mixture of constituents in aerosols(e.g.,losses of semi-volatile organic com-pounds and ammonium nitrate may occur,while,by contrast,particle-bound water may be retained and vary according to the seasons and sampling location [9]).As no standard suspended aerosol exists,inter-comparison of the di?erent RMs on the same atmos-pheric sample would be the only way to evaluate the comparability of the di?erent devices.Acritical review on the measurement of suspended particles can be con-sulted for more details[10].

4.3.Need for technology development

Quality objectives￠xed by regulation are sometimes too ambitious with respect to the state of knowledge and technical development(e.g.,Demerjian[6]noticed that for PM2.5in USA,for which quality objectives were sta-ted in US EPA40CFR Parts53and58,1997,consider-able development work remains to be done before routine monitoring can be achieved).

4.4.Summary

These di?erent examples show that,even if RMs and QA/QC procedures are well de￠ned,ensuring trace-ability remains a continuous challenging task invol-ving the improvement of the existing instrumentation for a better routine operation,the continuous develop-ment of reliable technologies to obtain representative measurements and to respond to new requirements in quality

objectives.

https://www.wendangku.net/doc/b9861860.html,/locate/trac255

5.Calibr ation

Calibration is a key step in the analytical process,and is therefore an important element in determining the accuracy,the precision and the robustness of analy-zers.For air analysis,calibration is generally performed by using gas conditioned in pressure cylinders at var-ious concentrations covering the range of interest.For routine use,working standards should be traceable to an NBS(National Bureau of Standards)SRM(Standard Reference Material).Certi￠ed gases are available from suppliers with NBS traceability documentation.An international standard(ISO6141:2000)describes the requirements for certi￠cates for calibration gases and gas mixtures.For zero checks,zero-grade nitrogen can be used.

Generally,a standard gas contains only one com-pound at a￠xed concentration.For VOC measure-ments,there is usually only one calibration curve for all compounds(e.g.,when a￡ame ionization detector is used and linear and cyclic alcanes are being calibrated). Indeed,in this case,the detector response is propor-tional to the e?ective carbon number in the molecule [11].It is thus possible to calculate the response coe?-cient for each alcane by using a reference propane stan-dard,as is recommended for total VOC determination in industrial emissions[12].

For other VOC measurements,standard gas mixtures are proposed by the suppliers.Since1990,these stan-dard mixtures are certi￠ed by inter-comparison exer-cises between di?erent international organizations [13].However,some stability issues are sometimes observed with regard to the compounds and their con-centrations in the mixture[14].Moreover,pressure cylinder mixtures are not relevant for checking the method performance,as it is not possible to vary the relative ratios between compound concentrations.

Di?erent alternative normalized methods for generat-ing standard gases are available.They are based on the static or dynamic generation modes described below. 5.1.Gravimetric methods

These methods are convenient when small volumes of standard gas are required.They are based on the intro-duction of a known mass of the target compounds (liquid or gas)in an known volume of dilution gas(air for example).Through successive dilutions,it is possi-ble to reach ppmv[15]or ppbv[16].This procedure is identical to that applied to pressure cylinders,and the international standard ISO/DIS6142:1999describes the criteria for ensuring the validity of the gas mixture.

5.2.Dynamic volumetric methods

These methods consist of introducing a￡ow-rate of gas A(pure compound or mixture)in a constant￡ow-rate of complementary gas https://www.wendangku.net/doc/b9861860.html,pared with gravimetric methods,they are particularly convenient when high sample volumes are required and when the in￡uence of air velocity should be checked on a sampling device. International standards(ISO6145/1to ISO6145/8) describe the numerous techniques that can be involved for the preparation of gas mixtures.Among them,per-meation is one of the most used.

5.2.1.Permeation.The principle involves the permea-tion of the target compound through a convenient membrane into a complementary gas￡ow.The target compound is placed in a tube which is maintained at constant temperature in a thermostatted chamber.The methodology is described in international standards ISO/DIS6145/10:1999and ISO6349:1979(e.g.,it is recommended for SO2,NO2,and benzene for concentra-tions ranging from10à6to10à9in molar fraction with a relative uncertainty of 2.5%(ISO/DIS6145/ 10:1999)).The permeation rate through the mem-brane depends on:the compound itself;the membrane structure;the temperature;and,the partial pressure of the compound in and out the tube.These parameters can be maintained constant when the system is used correctly.

Asource of errors can be the experimental determina-tion of the permeation rates,which requires precise weighing.To overcome this drawback,it is possible to buy permeation tubes certi￠ed by suppliers for permea-tion rate,uncertainty and lifetime.

This method is therefore one of the most convenient for generating gas mixtures because of its robustness and precision.However,the system requires high equi-libration time,may not be cost-e?ective for complex mixtures(as1tube is required per compound)and moreover,it is not possible to modify the relative concentrations of compounds,which depend on their permeation rates.

An example of home-made system using the permea-tion method for generation of trace sulfur compounds is shown in Fig.1[17].Successive dilutions of

the

256https://www.wendangku.net/doc/b9861860.html,/locate/trac

standard gas generated were carried out by using mass-￡ow meters calibrated for air by the supplier[17,18]. These￡ow meters were preferred to rotameters,as they are more precise:1%of the maximal￡ow rate.It should be noted that this precision can be obtained only if a pressure drop of1bar applies between the inlet and the outlet of the mass-￡ow meter[19].To improve homo-geneity,a mixing chamber was introduced.It is also possible to check the in￡uence of air humidity on the analytical method by introducing an humidi￠cation system[17].The presented set-up allowed each of the ￠ve sulfur compounds studied to be generated in the concentration range0.01^1000 g/m3[17].

5.2.2.Syringe injection.This method is particularly recommended when calibration for numerous com-pounds in a mixture is required.By contrast with pres-sure cylinders and permeation tubes,it is possible to modify easily the composition of the standard mixture qualitatively and quantitatively[19].However,this method is suitable only for the generation of atmos-pheres of gases or low-to-medium boiling liquids(up to 140 C)[20].

The principle involves loading organic gases or liquids into a gas-tight syringe connected to an atomi-zation chamber.Injection in a complementary gas(air)￡ow stream is carried out by means of a motor-drive on the syringe.This method is recommended for con-centrations ranging from ppmv to%with dilution gas ￡ow rate in the range1^100L/min(ISO6145/4: 1986).If desired,the atmosphere generated can be fur-ther diluted by using a set-up similar to that presented in Fig.1.

Ahome-made device based on syringe injection is shown in Fig.2.This system allows the generation of gas mixtures containing more than10VOCs[21].The concentration range obtained at the￠rst dilution step is 1^$100mg/m3with a precision of1%[19].By means of the second dilution step,it is possible to generate sub- g/m3concentrations with a precision of10% [19],which is particularly useful for checking the per-formance of new pre-concentration techniques for VOCs,such as SPME(solid-phase microextraction)[21^ 23].

Three major sources of errors can be identi￠ed with the syringe-injection technique:

atomization may not be quantitative for rela-tively high boiling compounds;

the motor-drive is expected to have a precision of 1%;and,

the￡ow meters are expected to have a precision of at least2%.

Moreover,the stability of the delivered concentra-tions should be carefully examined because of slight irregularities in the syringe drive[20].

5.3.Summary

To summarize,a lot of techniques are available for generating standard gases.In most cases,certi￠ed pres-sure cylinders will be preferred,but some speci￠c appli-cations need more￡exibility(e.g.,complex mixtures and low concentrations),so the analyst should make his own selection to determine the best￠t between his analytical objectives and the performance of the system chosen.

As noted above,numerous methods and principles are described in international standards,but few have been commercialized,so home-made devices are needed.To ensure good accuracy,precision and trace-ability from one laboratory to another,validation should be carried out through comparison with a reference system by following the procedure described in ISO/DIS6143:1998.

6.VOCs as a case study

VOCs include non-methane hydrocarbons(NMHCs) and oxygenated organic compounds that play an important role in the photochemistry of the troposphere.Some of these compounds,such as ben-zene,are carcinogenic,leading to a limit value￠xed to protect human health of5 g/m3(see Section3above). Di?erent analytical methodologies can be applied for VOC monitoring in ambient air,and standards describ-ing RMs are listed in Table2.These methodologies are based on gas chromatography(GC)equipped with FID (￡ame ionization detector),PID(photo ionization detec-tor)or mass spectrometer detector(MSD).

6.1.RMs

On-site air sampling can be performed on-line with GC for automated￠eld measurements or o?-line with GC analysis at laboratory.

For on-line analysis,sampling generally involves pre-concentration of VOCs by adsorption on solid materials. Di?erent devices are used,depending on the di?erent analytical objectives:

Figure2.Standard gas-generation device for trace VOCs based on the

https://www.wendangku.net/doc/b9861860.html,/locate/trac257

adsorbent tubes for dynamic sampling(air is pumped through the adsorbent bed)are recom-

mended for several minutes to several hours

average sampling times;

whereas di?usive sampling cartridges,based on the uptake of VOCs following Fick’s First Law,

are suitable for long-term exposure evaluation

(several days to several weeks).

For both methods,many di?erent adsorbents with di?erent adsorption e?ciencies are available(e.g.,acti-vated charcoal presents the highest adsorption capacity but desorption for analysis is carried out only by solvent extraction,generally by using harmful carbon disul￠de [24,25]).The solvent is then injected in a standard GC con￠guration.However,several mL are needed to com-plete quantitative extraction but only1 L is injected, so there is a loss of sensitivity.Moreover,this desorption

method is not easy to automate,so it requires a manual procedure in the laboratory.

More recently,thermal desorption has been devel-oped to provide better sensitivity and more reliability for routine operation in situ.The main disadvantage of this technique against solvent extraction is the instru-ment cost,as a speci￠c automated thermal desorber (ATD)is needed.For this technique,weaker adsorbents than activated charcoal are required[25].Most used for VOCs are organic polymers(e.g.,Tenax TA)and carbon-based materials,such as graphitized carbons (e.g.,Carbopack B)or micro-porous solids(e.g., Carboxen and Carbotrap).Detailed reviews and method comparisons are presented by Harper[26],and Skov et al.[25].

Air sampling without pre-concentration is also applied by using canisters that are particularly adapted to the most volatile and non-reactive VOCs[25].For trace analysis,the inner surface of canister should be inert and covered by deactivated fused silica named Silcosteel[27].Under these conditions,it is possible to store various VOCs at ppb level up to30days after sampling[28].However,this material is expensive and its cleaning is not easy(a speci￠c device being required),leading to possible cross-contamination from one sample to another.By using this sampling mode, pre-concentration and analysis are carried out in the laboratory[25]and only grab sampling can be carried out.

Fig.3resumes the di?erent standardized sampling methods for VOC monitoring and shows the complexity to make the correct choice from among the di?erent RMs proposed.

https://www.wendangku.net/doc/b9861860.html,parison of methods

The choice of method is of course also related to the information required.For example,because health e?ects from benzene are based on cumulative exposure,long sampling times seem to give the most pertinent information.For other VOCs,short sampling times would be preferred to identify pollution peaks and cor-relate them to meteorological parameters and ozone concentration.Whatever the sampling technique used, results should be traceable to be compared,so compara-tive studies have been carried out.For benzene,Skov et al.[25]have shown that a di?usive sampler measures 20%higher values than a BTEX(benzene,toluene, ethylbenzene,xylene)monitor based on continuous measurement by automated GC.This systematic deviation may be because of the uptake rates used for di?usive sampler[25].Another source of error can be related to the di?erence in the concentration evaluated: a calculated weekly average for BTEX monitor;and,a measured average for di?usive sampling.By contrast, comparison of a BTEX monitor and dynamic sampling on adsorbent tube for1hour showed no signi￠cant di?erence[25].

For VOC analysis,Czaplicka et al.[29]have com-pared sampling on charcoal tube followed by CS2 extraction and sampling on Carbotrap followed by automated thermodesorption.Standard deviations for the￠rst technique are more than10times above those of thermal desorption for BTEX,and detection limits are also high(around1.2 g/m3against0.4 g/m3for thermal desorption)[29].Moreover,￠nal results obtained with solvent extraction can be easily a?ected by errors produced by the operator and also by CS2, which can be very easily contaminated by VOC vapors from air.The extract should be therefore stored under inert gas[29].By contrast,automated thermodesorp-tion avoids contamination,losses because of solvent evaporation and errors because of manual operation. For these reasons and also because on-site methodolo-gies are now preferred for ambient air monitoring, methods involving solvent extraction tend to have been progressively

abandoned.

Generally,further e?orts should be made to explain the discrepancies found between methods.With this aim,inter-comparison programs,such as NOMHICE (non-methane hydrocarbon inter-comparison experi-ment)[30]and AMOHA(accurate measurements of hydrocarbons in air)[25,31]were implemented to assess the quality of measurements.

6.3.Sampling and analytical variability

For any given methodology,di?erent sources of result variability can be determined.These causes of uncer-tainties are taken into account in QA/QC procedures. These uncertainties may arise from[31]:

Unreliable standards:for NMHCs,daily calibra-tion is achieved through the e?ective carbon

number concept described in Section5,and zero

gas measurements are also carried out regularly

for blank determination.

Loss,contamination and reactions:to avoid loss or reactions in the analytical system,inactivated

Silcosteel valves and tubings are used.However,

another source of sample modi￠cation has been

attributed to the Na￠on membrane that is used

before the adsorbent trap to remove water vapor.

Even if this technique is more selective than

other common dryers,such as magnesium chlo-

rate[31],some polar VOCs(e.g.,aldehydes and

ketones)are signi￠cantly retained and artifacts

because of alkene rearrangement during ther-

mal regeneration of Na￠on are observed[32].

Therefore,this dryer can be used only when non-

polar VOCs are investigated,such as NMHCs,

and a speci￠c test should be implemented in the

QA/QC procedures.

Misidenti￠cation and erroneous quanti￠cation: these occur because of chromatographic peak

overlap and bad baseline.Peak overlap is often

because of the complexity of ambient air matri-

ces.To simplify the sample,Na￠on dryer is used

to remove polar compounds that may be co-

eluted with non-polar VOCs.In some cases,

manual peak integration should replace auto-

matic mode.

Bortnick et al.have studied the relative contribution of sampling,analysis and environmental parameters to monitoring variability[33].The relative importance of sampling and analytical errors varies by compounds: for carbonyls,sampling appears a more important source of error,whereas for several toxic VOCs(ben-zene,carbon tetrachloride,methylene chloride),analy-tical error is predominant.Generally,they also stated that errors caused by environmental variability(spatial and temporal)are more important than monitoring errors[33].

By contrast,for low concentrations(<0.5ppb), monitoring uncertainties are predominant[33].In this particular case,low concentrations probably represent the background pollution and should not be sig-ni￠cantly in￡uenced by sampling location,meteor-ological parameters or anthropogenic sources.

7.Conclusion

Air-monitoring networks have been implemented since the early1980s.Since then,considerable e?orts have been made to improve assessment of air quality through new regulations and development of SOPs and QA programs,including inter-comparison exercises.

For most of the target pollutants,RMs are well de￠ned and commercial monitoring instrumentation performs satisfactorily.With su?cient QA/QC,these RMs are expected to satisfy the primary objectives of the monitoring network.However,signi￠cant measure-ment issues remain to achieve the objective of quality data and to enhance the performance of the monitoring network.For example,new criteria for pollutants,such as VOCs or PM2.5,require considerable technological and analytical developments because of the lack of sensitivity,signi￠cant variability and the non-repre-sentative nature of the sample,especially at trace concentrations.

For high concentration levels,as stated by Bortnick et al.[33],it would be more interesting to increase the number of sampling locations and frequency rather than use sophisticated analytical procedures to arrive at a better understanding of spatial and temporal variabilities of data.

In this respect,atmospheric modeling is also envi-saged as a reliable tool for overcoming the poor spatial coverage of monitoring data and contributing to their traceability.

References

[1]US Environmental Protection Agency,Regional Approaches to

Improving Air Quality,EPA/451-K-97-001,US EPA, Washington,DC,USA,1997.

[2]J.S.G.D.Santos-Alves,R.F.Patier,Sens.Actuators B59(1999)

69.

[3]CITEPA,Methodology for calculating emissions into the air,

2001(https://www.wendangku.net/doc/b9861860.html,).

[4]R.N.Colvile,N.K.Wood￠eld,D.J.Carruthers,B.E.A.Fisher,

A.Rickards,S.Neville, A.Hugues,Environ.Sci.Policy5

(2002)207.

[5] C.Borrego, A.I.Miranda,M.Coutinho,J.Ferreira,

A.C.Carvahlo,Environ.Poll.120(2002)115.

[6]K.L.Demerjian,A tmos.Environ.34(2000)1861.

[7]http://www.gazettelabo.tm.fr/2002archives/publics/2000/

52LNE.html

[8]US Environmental Protection Agency,List of Designated Refer-

ence and Equivalent Methods,October1,1996.NERL,Air

https://www.wendangku.net/doc/b9861860.html,/locate/trac259

Measurements Research Division,Research Triangle Park,NC, USA,1996.

[9]J.C.Chow,J.P.Engelbrecht,N.C.G.Freeman,J.H.Hashim,

M.Jantunen,J.P.Michaud,S.S.de Tejada,J.G.Watson,F.Wei, W.E.Wilson,M.Yasuno,T.Zhu,Chemosphere49(2002)873.

[10]J.C.Chow,J.Air Waste Manage.Assoc.45(1995)20.

[11] C.Badol, A.Borbon,N.Locoge,T.Le onardis,J.C.Galloo,

J.Anal.Bioanal.Chem.(in press).

[12]AFNOR,‘‘Qualite de l’air-Emissions de sources￠xes-De termi-

nation d’un indice relatif aux compose s organiques en phase gazeuse-Me thode par ionisation de￡amme’’NF X43-301 (French Standard)AFNOR,Paris,France,1991.

[13]P.E.Holland,P.T.Woods,C.Brooks,Accred.Qual.Assur.6

(2001)120.

[14]R.Benesch,T.Jacksier,Anal.Chem.73(2001)379.

[15]M.Chai,J.Pawliszyn,Env.Sci.Technol.29(1995)693.

[16] A.Nielsen,S.Jonsson,J.Chromatogr.A963(2002)57.

[17] F.Lestremau, F.A.T.Andersson,V.Desauziers,J.L.Fanlo,

Anal.Chem.75(2003)2626.

[18] F.Lestremau,V.Desauziers,J.C.Roux,J.L.Fanlo,

J.Chromatogr.A999(2003)71.

[19]L.Tuduri,Ph.D.Thesis,Universite de Pau et des Pays de

l’Adour,France,2002.

[20]Occupational Medicine and Hygiene Laboratory,Method for

the determination of hazardous substances-Generation of test atmospheres of organic vapors by the syringe injection

technique,MDHS3,Health and Safety Executive,London,UK, 1990.

[21]L.Tuduri,V.Desauziers,J.L.Fanlo,J.Chromatogr.Sci.39

(2001)521.

[22]L.Tuduri,V.Desauziers,J.L.Fanlo,J.Microcolumn Sep.12

(2000)550.

[23]L.Tuduri,V.Desauziers,J.L.Fanlo,Analyst(Cambridge,U.K.)

128(2003)1028.

[24] E.Matisova,S.Ibrabukova,J.Chromatogr.A707(1995)145.

[25]H.Skov,A.L indskog,F.Palmgren,C.S.Christensen,A tm.

Environ.35(2001)S141.

[26]M.Harper,J.Chromatogr.A885(2000)129.

[27]M.Koga,Y.Hanada,J.Zhu,O.Nagafuchi,Microchem.J.68

(2001)257.

[28] D.W.M.Sin,Y.C.Wong,W.C.Sham, D.Wang,Analyst

(Cambridge,U.K.)126(2001)310.

[29]M.Czaplicka,K.Klejnowski,J.Chromatogr.A976(2002)

369.

[30] E.C.Apel,J.G.Calvert,T.M.Gilpin, F.C.Fehsenfeld,

D.D.Parrish,W.A.Lonneman,J.Geophys.Res.104(1999)

26069.

[31] C.Plass-Du lmer,K.Michl,R.Ruf,H.Berresheim,

J.Chromatogr.A953(2002)175.

[32]Q.Gong,K.L.Demerjian,J.A ir Waste Manage.A ssoc.45

(1995)490.

[33]S.M.Bortnick,S.L.Stetzer,Atm.Environ.36(2002)1783.

260https://www.wendangku.net/doc/b9861860.html,/locate/trac

测量溯源性控制程序

1 目的为使实验室使用的检测仪器设备溯源到国家计量基准，确保检测数据准确可靠，具有可比性，特制定本程序。 2 适用范围本程序适用于测量仪器溯源管理工作。 3 职责 3.1 技术负责人对实验室量值溯源的有效性负责。 3.2 设备员： 3.2.1 制定仪器设备检定计划并组织实施。 3.2.2 保存仪器设备溯源结果和证书。 3.2.3 负责检测室测量设备台帐的动态管理。 3.3 检测室质量负责人配合、支持、监督仪器设备的检定并组织实施对仪器、设备期间核查。 4 工作程序 4.1 测量仪器设备的量值溯源 4.1.1设备员对实验室的测量仪器设备进行动态管理，填写“监视和测量设备分类台账” 4.1.2 实验室设备员根据各检测仪器设备的检定周期提出定期检定计划，经技术负责人批准后，报公司设备管理部计量科，由其组织有资质的计量部门进行检定，并负责测量设备台帐的动态管理。检定计划中应包括实验室在用的所有出具检测数据的仪器、设备、量具（包括新购置的、修理过的等）。设备员应在检定有效期满前一个月联系安排下一周期的检定工作。 4.1.3承接检定的机构或部门的计量标准器具及配套设施，应是经有资质的计量部门检定（校准）合格的，并能溯源到上一级计量基准，直至国家基准和国际单位制。发布的检定证书应有符合确定的计量规范声明或包含测量结果的不确定度等相关内容。 4.1.4 设备员应对经检定的仪器设备、量具粘贴相应标识，其标识管理按《质检中心设备管理办法》执行。检定完成后，检定证书原件由实验室设备员存档。

4.1.5 对直接影响检测质量的、使用频繁的、漂移较大的、容易出现超载的仪器设备，可以定期进行期间核查，以证明其量值稳定可靠。 4.2 量值溯源关系图，见附表测量量值溯源关系图

临床检验的量值溯源问题

临床检验的量值溯源问题一、引言近年来，临床检验的量值溯源问题在国际上受到广泛重视。欧洲议会和理事会1998年10月签署一项将于2003年12月生效的关于体外诊断器具的指令(Directive 98/79/EC)[1]，该指令的一项关键内容是要求体外诊断器具的校准物质和/或质控物质定值的溯源性必须通过已有的高一级的参考方法和/或参考物质予以保证。欧洲指令是法律文件，生效后有关各方必须执行。为配合该欧洲指令的实施，国际标准化组织(ISO)于1999年起草了5个相关标准，其中与生产厂家关系比较密切的是ISO/DIS 17511“校准物质和质控物质定值的计量学溯源性”[2]和ISO/DIS 18153“酶催化浓度校准物质和质控物质定值的计量学溯源性”[3]。以上指令和标准主要针对诊断试剂的生产。对临床实验室检验来说，作为国际实验室认可依据的ISO/IEC 17025 “检测和校准实验室能力的通用要求”[4]（我国国家标准GB/T 15481-2000和国家实验室认可委员会CNACL 201-2001“实验室认可准则”等同采用ISO/IEC 17025）和ISO/FDIS 15189 “医学实验室质量管理”[5]也都对临床检验结果的溯源性作出明确要求。鉴于量值的溯源性将可能成为体外诊断试剂生产和使用中的重要质量指标，而我国试剂生产者和临床检验工作者对此计量学概念可能还不太熟悉，本文介绍临床检验量值溯源的基本概念、现状及有关问题。二、量值的溯源性及溯源链的结构和工作原理 ISO对溯源性的定义如下：测量结果或标准量值的属性，它使测量结果或标准量值通过连续的比较链与给定的参考标准联系起来，给定的参考标准通常是国家或国际标准，比较链中的每一步比较都有给定的不确定度。不确定度是另一个计量学术语，ISO对它的定义为：与测量结果相关的参数，代表可能可合理地赋予被测量的值的分散性。不确定度评定有A、B两类，A类评定基于测量结果，B类则基于经验或其它信息的概率分布，A类和B类合成为标准不确定度。临床检验的量值溯源可以有不同模式，但其中心内容是使各测量方法的测量值与一公认的标准发生联系。图1为ISO 17511[2]描述的溯源图。一个样品或参考物质的测量结果的溯源性通过一系列对比测量而建立，对比测量中的测量过程和校准物质的计量学等级由低到高组成一条连续的链(溯源链)。链的顶端是国际单位制(SI)单位(基本或导出单位)，SI单位国际通用，不随时间和空间的变化而变化，因此它们是溯源链的最高等级。一级参考测量过程是具有最高计量学特性的参考测量过程，它须是基于特异、无需同量校准而能溯源至SI单位、低不确定度的测量原理，目前认为可用于一级参考测量过程的测量原理仅限于同位素稀释/质谱(ID/MS)、库仑法、重量法、滴定法和依数性(如凝固点降低)测量等。一级参考物质是测量单位的体现体，具有最可能小的测量不确定度，它可由一级参考测量过程直接定值，也可通过可靠的杂质分析间接定值，一级参考物质一般是高度纯化的被测物质。二级参考测量过程是经充分论证，其不确定度能满足特定要求，能用于低一级测量过程评价和参考物质鉴定的测量过程，二级参考测量过程用一级参考物质校准。二级参考物质用一种或多种二级参考测量过程定值，一般具有与实际样品相同或相似的基质，主要用于量值传播。一级和二级参考测量过程的建立和维持及一级和二级参考物质的制备有高度的知识、技术和设备要求，故一般由国际或国家计量机构及经认证的参考实验室完成。一级和二级参考物质一般是经计量权

溯源性与可追溯性

1 溯源性与可追溯性的定义溯源性：指一个测量结果或测量标准的值，都能通过一条具有规定不确定度的连续比较链，与测量基准联系起来。可追溯性:一个过程的可追查的范围和程度。 2 可追溯性体外诊断生产实施细则中规定：“主要物料的采购资料应能够进行追溯，应当按照采购控制文件的要求保存供方的资质证明、采购合同或加工技术协议、采购发票、供方提供的产品质量证明、批进货检验（验收）报告或试样生产及检验报告。”这里的可追溯主要是指采购物料的来源去向及物料来源的合法性。还有如一次性医疗器械的采购、使用的可追溯性，生产过程的可追溯性（批生产记录的追溯）以及生产批号的可追溯性，规定了其可追溯的范围和程度。 3 溯源性体外诊断试剂生产实施细则中规定：“外购的标准品和质控品应能证明来源和溯源性。” 、“使用一级标准物质、二级校准物质应能够对量值进行溯源。对检测中使用的校准品和质控品应当建立台帐及使用记录。应记录其来源、批号、效期、溯源途径（或靶值转换方法）、主要技术指标、保存状态等信息。” 体外诊断试剂校准品、质控品研究技术指导原则中规定：“3.6校准品的溯源性、互换性，定值质控品赋值的统计学处理,应提供校准品的溯源性资料，计量学溯源链的说明应始于该产品的值，止于计量上最高参考标准。校准品如有互换性，应提供互换性研究资料---”。计量学溯源性（溯源链等级及校准传递方案） GB/T 21415-2008/ISO 17511：2003 体外诊断医疗器械生物样品中量的测量校准品和控制物质赋值的计量学溯源性于2008年9月开始实施。对校准品和控制物质赋值的计量学溯源类型及如何溯源规定的比较详细。目前,在临床实验室可测定约1500个不同的分析物，但是只有30个满足溯源性的要求，达到理想的终点。试剂盒中通常配置的阴阳性对照品，其实严格的来说只能算是试剂盒内对照，通过阴阳性对照品的实验结果可以提示该产品的该次实验是否成功以及可信度有多少。质控品：IFCC的定义是专门用于质量控制目的的标本或溶液，不能用于校准。对稳定性、瓶间差要求高。分定值和不定值两种。在选择控制品时，应该选择有几个浓度的、浓度范围分布较宽的、最好是医学决定水平的、有可报告范围范围是上下限值的质控品。标准品：用于定标即标准曲线的绘制。校准品：公司指定用来校准某检测系统（仪器+试剂+方法程序）的，是考虑到它具有基质效应的情况下，人为赋予校准品的校准值。因此，校准品必须专用于某一检测系统。标准品的定值一般而言，检验工作中使用的标准品属应用标准。将符合质量标准的纯品使用称量法和容量法配制成溶液。用决定性方法反复测定，结果在规定的范围内属合格。测定制的可靠性取决于鉴定方法，分析方法的可靠性不如公认的称量法和容量法。标准品值由称量和容量法计算确定。决不可将实测值替代修正。三、引入校准品

量值溯源要求

CNAS—CL06 量值溯源要求中国合格评定国家认可委员会二〇〇六年六月

量值溯源要求 1．前言 1.1量值溯源的一致性是国际间相互承认测量结果的前提条件，中国合格评定国家认可委员会（英文缩写：CNAS）将量值溯源视为测量结果可信性的基础，CNAS对量值溯源的要求与国际规范的相关要求一致。 1.2 在量值溯源方面，CNAS要求已认可机构在满足国际标准的同时应遵循我国有关法律法规的规定。 1.3 CNAS承认一切能够证实与国家或国际计量基（标）准存在有效关系的境内外量值溯源途径。 1.4 CNAS承认国家计量基（标）准。 2．适用范围本要求规定了量值溯源要求，适用于已认可机构从事的测量活动，也适用于CNAS对已认可机构量值溯源方面的评估和报告。 3. 引用文件下列文件中的条款通过引用而成为本文件的条款。以下引用的文件，注明日期的，仅适用引用的版本；未注明日期的，适用引用文件的最新版本（包括任何修订）。 3.1 《中华人民共和国计量法》（1986年7月1日施行） 3.2 《中华人民共和国计量法实施细则》（1987年2月1日发布） 3.3 International Vocabulary of Basic and General Terms in Metrology(VIM). BIPM, IEC, IFCC, ISO, IUPAC, IUPAP, OIML, 2nd edition, 1993. 《国际通用计量学基本术语》 3.4 JJF1001-1998《通用计量术语及定义》 3.5 ILAC—P10：2002《ILAC关于测量结果溯源性政策》 3.6 ISO/IEC 17011：2004《合格评定—认可机构通用要求》

可溯源性

解决溯源性是临床检验结果准确性的关键重庆医科大学陈宏础一个病人的苦恼：“为什么几家医院检查不同结果，我相信谁？现在检查很多，价钱都很贵，实在承受不了，无奈！希望遇到一个好大夫，把我的病诊断清楚，早日治好，这是全家最幸福的事”。 ISO/IEC17025的质量体系中就提出“按时、准确、客观地报告检验结果”。如何做准确？一、测量结果的准确性(accuracy)的新观念：准确性=正确性(trueness)+精密度(precision)+其他测量能力（线性范围、灵敏度、特异性）。现在人们十分重视：严格按规范化操作，SOP是检验科共同遵守的法规；临床检验量值必须校准。但如何校准？这就涉及溯源性问题。如何解决溯源性？二、溯源性的定义： “测量结果或标准的值通过连续的比较链与一定的参考标准相联系的属性。参考标准通常是国家或国际标准，比较链中的每一步比较都有给定的不确定度，此定义适用于各领域的测量。” 所以，溯源性是解决的测量结果的正确性（Trueness of Measurement），即测量均值与真值的一致程度。以往我们只认为我们的临床标本只有用国际上公认的决定性方法和

国际公认的标准品去测定，其结果才接近“真值”。如果没有这些条件怎么办？ ISO一系列文件回答了这个问题。如果有SI单位制单位，又有国际公认的原级参考程序(即有原（一）级参考测量程序，又有国际标准物质，)，其量值溯源性见图1 “连续比较链”是指“正确性”由低级到高级的、交替出现的测量程序和校准物，如厂家提供给临床实验室常规测量程序由同时提供的产品校准物校准，而产品校准物由高一级的测量程序定值，此高一级的测量程序由更高一级的校准物校准，以次类推，直至“一定的参考标准”，即国际公认的参考物质和参考测量程序。上述比较链又称溯源链。每一次结果传递均带来一定的不确定度。三、稿清楚日常遇到的一些概念问题

医学实验室的量值溯源性标准操作程序

SOP_04-1 医学实验室的量值溯源性标准操作程序一、目的：确保仪器正常运转与结果的准确性，严格检验质量标准，为临床提供及时、可靠的结果报告。二、适用范围：检验项目的量值溯源。三、操作人员：检验科授权工作人员四、操作步骤：溯源性(traceability)：通过一条具有规定不确定度的不间断的比较链，使测量结果或测量标准的值能够与规定的参考标准，通常是与国家标准或国际标准联系起来的特性。[VIM：1993，定义6.10]溯源顺序往往采用溯源等级图(也称溯源性体系表)来描述，用以表明测量仪器的计量特性与给定量的测量基准之间的关系，是溯源性的证据。建立溯源等级图的目的，是要对所进行的测量在其溯源到计量基准的途径中，尽可能减少环节和降低测量不确定度，并能给出适当的置信度。为实现溯源性，用等级图的方式应给出：对不同等级测量标准的选择；等级间的连接及其平行分支；有关测量标准特性的重要信息，如测量范围、准确度等级或最大允许误差等；溯源链中比较用的装置和方法。 ISO 15189：2003(E)文件中的5.6.3指出：应设计并实施一套对测量系统的校准，真实性的检定程序，以保证测量结果可溯源到SI单位，或可参比到一个自然常数或其他规定的参考值。如果以上都无法实现，还可采用参加适当的实验室间的比对计划；使用相应的参考物质(有证书说明其材料的特性)；将供应商或制造商提供的关于试剂、程序或检验系统的溯源性的说明形成文件等。 “不间断的比较链”在医学实验室的化学测量中，它是指通过不同级别的测量程序，参考物质和校正物而实现的连续测量，用一个测量程序为某种物质定值，该物质用做下一级测量程序的校准物，依此类推。比较链也称溯源链。 “规定的参考标准”为参考测量程序和参考物质。与参考标准的联系可以是直接的，也可以通过中间测量程序和校准物间接进行，即溯源链可长可短，但理论上应使溯源链尽可能地短。 “不确定度”是评价溯源性的指标。检测系统与参考系统的联系有好有坏，故在给出溯源性时也要溯源性过程每一步的不确定度，以说明最终结果与参考系统联系的质量。溯源等级图应注意区别测量标准本身的复现量值的不确定度，以及经该测量标准校准所得结果的不确定度。对于普通等级的测量仪器，也可以指出其最大允许误差。等级图中所反映的信息，应与有关的法规、规程或规范的要求相一致。 1 可溯源至SI单位的分析物量值溯源： 1.1 可溯源到SI的分析物：物质结构清楚、均一(纯)、有明确分子量的，试验结果以 mol／L 表示的物质称分析物(analyte)。大约有100个临床化学项目可以应用参考测量系统，追踪到SI的分析参考物参考测量系统注：(+)指潜在可利用性

测量溯源性

第一讲：测量溯源性众所周知，在许多领域中，分析测量的结果愈来愈多地用作为重要的决策依据。上至国家领导人制定政策，下到普通百姓日常生活，都与分析测量的结果密不可分。如：百姓每天都在关心的空气质量问题、食品的安全问题和医疗保健问题（体检的各项指标）；企业家关心国际贸易结算（天然气贸易）、关系到企业存亡的污染物排放问题；科技工作者关心物质的成份和材料的特性；国家领导人关心实施可持续发展战略及温室效应产生的全球气候变暖问题等等；对于以上所有问题判断和决策都离不开分析测量。分析测量涉及到科技、经济和社会发展的各个领域，很小的分析失误都可能导致产生长远影响的错误决策。因此，分析测量的结果可靠性、有效性至关重要。科技、经济和社会发展的各个领域对分析测量存在着巨大的需求，现有分析科技资源仅仅满足了其中的一部分。分析测量资源是一种稀缺资源。但是，这种稀缺的资源并没有很好的得到有效的利用。在英国，每年都要进行数百万次分析测量，尽管己采取了一些措施，但仍有较大比例的分析测量“未能满足用途”要求。美国每年用于医疗保健的消费约为1万多亿美元，约占GDP（国内生产总值）的13%，在这些开支中有超过20%是花在分析测量上。据估计，这些测量有三分之一属于无效测量。我国是一个发展中大国，分析测量的科技资源更是紧缺。但是，临床检测结果不能跨医院使用，异地测量结果可比性差等现象严重存在，分析测量稀缺资源没有得到有效的利用。分析测量有效性亟待提高。全球经济一体化己成为当今世界发展不可逆转的一大潮流。加入WTO标志着我国将更快地融汇到世界经济圈中去，全球化趋势必将对我国的经济、科技、文化及社会生活带来深刻地影响。一国在这个资源全球化配置和世界新经济秩序的建立过程中能获得多少利益，有多少发言权，测量结果是否能够得到多边采纳，很大程度上取决于本国的国家测量能力，例如涉及TBT（贸易技术壁垒）的技术法规与技术标准的建立与实施，依据的就是本国的测量技术实力。测量能力愈来愈明显地成为国家竞争力的重要组成部分，有效分析测量也愈来愈多地成为维护国家经济利益和保护国家经济安全的重要技术手段。因此，不断提高分析测量结果的质量，提高分析结果的有效性，具有十分重要的意义。实现有效性的重要措施之一就是努力实现分析测量向国际单位（SI单位）或公认单位的溯源。 1. 古代测量的溯源可比性和溯源性并不是一个新的概念。早在公元前3000年，古埃及人在修建金字塔时就利用了这一概念来保证测量结果的质量。腕尺定义为法老的前臂加上手掌的长度。用一根木棒来复现腕尺，就成了长度测量中保证可比性的第一个工作标准，在金字塔建造现场使用。后来，埃及人又用花岗岩做成更耐用的“皇家主腕尺”作为基准。在每个满月日，负责工作标准的人就会去“皇家主腕尺”处，将“标准尺”与这个“基准尺”比较，以确保“标准尺”的准确度。那个时代的金字塔就是利用这种可溯源的测量保证体系建造起来的。长度测量的一致性满足了测量目的的要求，在230米的距离上相对准确度达到了0.05%。 2. 测量溯源的定义

CNAS-CL0620xx《测量结果的溯源性要求》

CNAS-CL06:20xx《测量结果的溯源性要求》修订说明本次修订主要是由于ILAC-P10《ILAC关于测量结果溯源性的政策》于2013年1月发布实施，针对ILAC-P10对以溯源为目的的校准服务的资格和能力的有关要求，需要修订CNAS-CL06:2011，以确保CNAS的溯源性政策以及获认可机构的测量结果的溯源性满足ILAC-P10的要求。本文件中的“注”给出了实施的途径或必要的说明，CNAS不再另行制定本文件的实施指南。CNAS将根据不同领域的专业技术特点，需要时，制定特定领域的量值溯源要求或指南。主要修订内容和变化： ●由于ISO/IEC指南99（VIM）已将“测量溯源性”术语修改为“计量溯源性”，本文件中使用“计量溯源性”术语； ●增加了计量溯源性、RM、NMI等的术语和定义； ●根据ILAC-P10:2013提出的新要求，明确了作为有效计量溯源性证明的证书中应包含溯源性和测量不确定度信息； ●明确了CIPM MRA、ILAC MRA和法定计量机构三个提供有效溯源服务的体系，以及其溯源证据满足ILAC要求的措施。对CIPM MRA所涵盖的国家计量院的溯源地位和范围做了详细的描述。增加了提供溯源服务的法定计量机构的范围。 ●加强了对获认可的校准实验室的溯源地位； ●不区分境内境外，溯源要求均一致；

●对特殊情况的溯源，与CL01的原则一致； ●进一步明确了标准物质/标准样品（RM）选用的要求； ●与本文件相关的CNAS-GL04:2011《量值溯源要求的实施指南》由于部分内容不符合ILAC溯源性政策，CNAS已于2012年发文废止。本次对CNAS-CL06:2011的修订，对部分指南性内容以注的形式给出，不再针对该文件制定指南文件，特殊领域的溯源性指南需要时另行制定。 CNAS 认可二处 2013年5月22日

临床检验量值溯源与参考系统讲解

?专家论坛?临床检验量值溯源与参考系统陈文祥【摘要】医疗卫生工作需要准确可比的检验结果,实现临床检验标准化的有效途径是建立和保证检验结果的计量学溯源性。溯源性通过一系列由不同等级测量程序和校准物实现的比较测量而建立, 较低级别程序足够特异和校准物具有互通性是计量学溯源的前提。较高级别的测量程序和校准物称参考系统,参考系统需符合特定计量学要求,为此有关国际组织成立检验医学溯源联合委员会(J CTL M, 鉴定现有国际参考测量程序、参考物质和参考实验室。总之,临床检验量值溯源问题已受到广泛重视, 发展和应用参考系统将可能成为临床检验领域的重要课题。【关键词】溯源性; Metrological traceability and reference systems in laboratory medicine C H EN Wen2x iang.B ei j ing Hos pital I nstitute of Geriat rics and N ational Center f or Clinical L aboratories B ei j ing100730,China 【Abstract】In laboratory medicine,accurate and comparable measurement results are required for correct medical interpretations1Metrological traceability is an important approach to achieve

standardization of test results obtained with different analytic systems1Traceability is established by a series of comparative measurements using a hierarchy measurement procedures and calibrators1 Specificity of the lower order methods and commutability of the calibrators are the prerequisites in establishing traceability1The higher order procedures and calibrators in the hierarchy are called reference systems and it is important that the reference systems are metrologically qualified1The Joint Committee for Traceability in Laboratory Medicine(J CTL Mhas been formed to identify the reference systems worldwide1In conclusion,the concept of metrological traceability has been introduced to the filed of laboratory medicine1Development and implementation of reference systems in laboratory medicine are expected1 【K ey w ords】Traceability; Reference system 临床检验结果准确,具有跨时空的可比性,是防病治病的需要,也一直是检验医学界的工作目标。实现检验结果准确的重要手段是建立和保证检验结果的溯源性,而开展检验量值溯源的必要条件是具备参考系统。作者介绍临床检验量值溯源有关问题和临床检验参考系统国际现状。

解决溯源性是临床检验结果准确性的关键

解决溯源性是临床检验结果准确性的关键一个病人的苦恼：“为什么几家医院检查不同结果，我相信谁？现在检查很多，价钱都很贵，实在承受不了，无奈！希望遇到一个好大夫，把我的病诊断清楚，早日治好，这是全家最幸福的事”。 ISO/IEC17025的质量体系中就提出“按时、准确、客观地报告检验结果”。如何做准确？一、测量结果的准确性(accuracy)的新观念：准确性=正确性(trueness)+精密度(precision)+其他测量能力（线性范围、灵敏度、特异性）。现在人们十分重视：严格按规范化操作，SOP是检验科共同遵守的法规；临床检验量值必须校准。但如何校准？这就涉及溯源性问题。如何解决溯源性？二、溯源性的定义： “测量结果或标准的值通过连续的比较链与一定的参考标准相联系的属性。参考标准通常是国家或国际标准，比较链中的每一步比较都有给定的不确定度，此定义适用于各领域的测量。” 所以，溯源性是解决的测量结果的正确性（Trueness of Measurement），即测量均值与真值的一致程度。以往我们只认为我们的临床标本只有用国际上公认的决定性方法和国际公认的标准品去测定，其结果才接近“真值”。如果没有这些条件怎么办？ ISO一系列文件回答了这个问题。如果有SI单位制单位，又有国际公认的原级参考程序(即有原（一）级参考测量程序，又有国际标准物质，)，其量值溯源性见图1 。 “连续比较链”是指“正确性”由低级到高级的、交替出现的测量程序和校准物，如厂家提供给临床实验室常规测量程序由同时提供的产品校准物校准，而产品校准物由高一级的测量程序定值，此高一级的测量程序由更高一级的校准物校准，以次类推，直至“一定的参考标准”，即国际公认的参考物质和参考测量程序。上述比较链又称溯源链。每一次结果传递均带来一定的不确定度。三、稿清楚日常遇到的一些概念问题临床上对一些概念必需搞清楚，如测定方法、标准物质、检测（分析）系统等。测定方法有：决定性方法（definitive method）：经详细研究尚未发现不准确度或不确定性原因的方法。参考方法（reference method）：经详细研究证明其不准确度与不精密度可以忽略的方法。常规方法（routine method）：可满足临床需要的日常使用的方法。我们对一些项目的常规方法比较了解，但对该项目的上一级方法未必清楚吧。对于标准物质、校准物、质控物应有充分的认识。

CNAS-CL06：2014《测量结果的溯源性要求》

CNAS-CL06 测量结果的溯源性要求Requirements on the Traceability of Measurement Results 中国合格评定国家认可委员会

前言计量溯源性是国际间相互承认测量结果的前提条件，中国合格评定国家认可委员会（CNAS）将计量溯源性视为测量结果有效性的基础，并确保获认可的测量活动的计量溯源性满足国际规范的要求。本次修订主要依据ILAC-P10:2013《ILAC关于测量结果溯源性的政策》，并结合我国计量溯源体系情况，规定了CNAS对测量结果的溯源性要求。本文件中的“注”给出了实施的途径或必要的说明，CNAS不再另行制定本文件的实施指南。需要时，CNAS将根据不同领域的专业技术特点，制定特定领域的量值溯源要求或指南。此次修订的主要变化如下： ●由于ISO/IEC指南99（VIM）已将“测量溯源性”术语修改为“计量溯源性”，所以本文件使用术语“计量溯源性”； ●根据ILAC-P10:2013提出的新要求，本文件明确了作为有效计量溯源性证明的证书中应包含溯源性和测量不确定度信息； ●进一步明确了标准物质/标准样品（RM）选用的要求； ●为与ILAC-P10:2013一致，文件名称改为《测量结果的溯源性要求》。本文件替代CNAS-CL06:2011《量值溯源要求》。

测量结果的溯源性要求 1．范围本文件规定了CNAS在对检测实验室（含医学实验室）、校准实验室、检验机构、标准物质/标准样品生产者和能力验证提供者等机构（以下统称合格评定机构）实施认可活动时涉及的测量结果的计量溯源性要求。本文件适用于检测（含医学检验）、校准活动，也适用于检验、标准物质/标准样品生产和能力验证等涉及测量活动的合格评定活动。本文件只适用于合格评定机构通过外部机构获得的测量结果的溯源性要求。合格评定机构实施的内部校准应满足CNAS-CL31《内部校准要求》。 2. 规范性引用文件下列文件中对于本文件的应用是必不可少的。凡是注日期的引用文件仅注日期的版本适用于本文件。凡是不注日期的引用文件，其最新版本（包括修改单）适用于本文件。 2.1 CNAS-CL01:2006 检测和校准实验室能力认可准则 2.2 ILAC-P10:2013 ILAC关于测量结果溯源性的政策 2.3 ISO/IEC指南99 国际计量学词汇—基础和通用概念及相关术语（VIM） 3. 术语和定义本文件采用VIM中的有关术语和定义。 3.1计量溯源性（VIM 2.41）通过文件规定的不间断的校准链，将测量结果与参照对象联系起来的测量结果的特性，校准链中的每项校准均会引入测量不确定度。注1：本定义中的参照对象可以是实际实现的测量单位的定义、或包括非序量测量单位的测量程序、或测量标准。注2：计量溯源性要求建立校准等级序列。注3：参照对象的技术规范必须包括在建立校准等级序列时使用该参照对象的时间，以及关于该参照对象的计量信息，如在这个校准等级序列中进行第一次校准的时间。

6.5计量溯源性

第6.5章计量溯源性 1总则检测中心根据检测方法和检测结果的溯源性要求，通过选择有资质有能力的校准实验室，以形成文件的不间断的校准链将测量结果与适当参考标准相关联，建立并保持测量结果的计量溯源性，其中每次校准对测量不确定度均有贡献。 2职责 2.1检测设备的检定校准：技术负责人负责管理；综合组负责实施； 2.2检测设备的期间核查：各检测组；测量溯源性的管理:技术负责人。 3范围适用于本中心所有与检测质量有关的计量仪器设备校准。 4内容 4.1检测中心通过以下方式确保测量结果可溯源到国际单位制（SI） a)选择有资质和具备能力的实验室提供的校准； b)选择具备能力的标准物质生产者提供并声明计量溯源至SI的有证标准物质的标准值； c)SI单位的直接复现，并通过直接或间接与国家或国际标准比对来保证。 4.1.1当技术上不可能计量溯源到SI单位时，检测中心通过下列方式证明可溯源至适当的参考标准： a)具备能力的标准物质生产者提供的有证标准物质的标准值； b)使用参考测量程序、规定方法或描述清晰的协议标准，其测量结果满足与其用途，并通过适当比对予以保证。 4.1.2检测中心制定《计量溯源性程序》，控制检测结果的溯源性。 4.2检定证书、校准结果的确认与管理 4.2.1技术负责人应对检定证书、校准报告或比对试验结果的确认进行审核，判定设备的测量准确度等技术性能是否符合检测工作的要求。 4.2.2当检定、校准出现超差或不合格，或比对及验证结果不满意时，设备使用人应对以前使用该设备所做的测量进行检查，确认是否对以前测量结果造

成了影响。如已造成影响则应对结果进行追踪，予以纠正。 4.3对于具有测量功能的检测设备，除非已证实校准带来的贡献对检测结果总的不确定度几乎没有影响。这种情况下，实验室应确保所用的设备能够提供所需的测量不确定度。 4.4量值溯源性的特殊（CNAS-CL01-A002:2018）要求 4.4.1总则的特殊要求 a)本中心应按检测方法的要求（当检测方法有要求时）建立校准曲线。所用标样应覆盖被测样品的浓度范围。最低浓度的标样应在接近检测方法报告限的水平,并应建立和执行线性校准曲线相关系数的准则。对非线性校准函数，需要更多的校准标样。如适用，应使用插入法技术(bracketing technique)。本中心至少使用5个标样(除空白外)建立线性校准曲线。更多的信息参考 ISO 11095“使用标准物质(参考物质)的线性校准”所给出的指南。 b)本中心应定期使用中间点的校准标样检查校准曲线，建立定期检查结果可否接受的判据，且该判据应与测量不确定度相当。本中心此类检查的频率取决于设备或方法的稳定性。通常情况下，约5％的检查频率就足够了，除非检测方法有其他要求，或设备极为稳定时可降低检查的频率。 4.6.2标准物质在使用期间应进行期间核查，核查可根据检测工作的实际，从标准物质的性状是否有异常变化、特性值是否稳定、储存环境是否符合要求等等方面着手。如果标准物质在期间核查中发现已经发生分解、产生异构体、浓度降低等特性变化，应立即停止使用，并追溯对之前检测结果的影响，执行《不符合检测工作控制程序》。 4.6.3本中心制定《试剂和标准溶液管理程序》，规定标准溶液和其他内部标准物质的制备、标定、验证、有效期限、注意事项或危害、制备人、标识等要求，并保存详细记录。标准溶液的配制应有逐级稀释记录。本中心标准溶液的标定应按照检测方法的要求或参照GB/T 601《化学试剂

检验结果量值溯源管理程序

检验结果量值溯源管理程序目的：建立和实施检验结果的量值溯源程序，使病人标本的测量结果能够通过一条具有规定不确定度的连续比较链，或通过实验室间的比对等方式，与测量基准联系起来，从而使测量结果的准确性得到技术保证。 2、范围：适用于检验科开展的检验项目。 3、职责： 3.1 各专业组长负责本专业组检验项目校准计划的制定以及室间比对计划、自建检测系统校准计划的制定和实施。 3.2 技术负责人负责校准计划、室间比对计划、自建检测系统校准计划的审批。 3.3 检验人员负责检验仪器的校准、比对项目标本的准备和检测。 4 定义和术语 4.1 量值溯源是指测量结果或测量标准的值，能够通过一条具有规定不确定度的连续比较链，与测量基准联系起来，从而使测量结果的准确性和一致性得到技术保证。 4.2 实验室间比对指按照预先规定的条件，由两个或多个实验室对相同或类似的被测物品进行检测的组织、实施和评价。 4.3 实验室内部比对

指按照预先规定的条件，由检验科内部实施的同一检验项目不同检测系统之间的比对，也包括各自实验室内部同一检验项目不同检测系统之间的比对。 4.4 方法学比较实验室准备用一个新的检测系统或测定方法，或新的试剂盒、新的仪器进行病人标本测定前，应与原有的检测系统或公认的参考方法一起检测一批病人标本，以评价新的检测系统或方法引人后的偏倚，从而决定其能否应用于临床；也包括不同检测系统之间所进行的比对。 5 工作程序 5.1 如果无法实现检验结果的量值溯源或不相关，还可采用以下方法（但不限于此）以提供结果的可信度： a ）对检测系统定期进行校准； b ）参加适当的实验室间比对活动； c ）使用相应的参考物质：此参考物质必须是有资格的供应商提供的有证标准物质，并附有材料特性的详细说明； d ）比率或倒易型测量； e ）使用已明确建立的、经规定的、性能已确定的、被各方承认的协议标准或方法； f ）利用供应商或制造商提供的试剂、程序或检测系统对溯源性的说明，形成实验室的溯源性文件 5.2 检测系统的校准

关于计量溯源性、校准间隔与期间核查

关于计量溯源性、校准间隔与期间核查 1溯源性是计量的基本特性计量是关于测量的科学，是实现单位统一、量值准确可靠的活动。计量的特点可以概括地归纳为准确性、一致性、溯源性和法制性四个方面。准确性是指测量结果与被测量真值的一致程度。由于实际上不存在完全准确无误的测量，与测量结果相联系的必然是反映测量质量(品质)的、适用于应用目的或实际需要的不确定度。从这个意义上说，计量是与测量结果置信度有关的、与不确定度评定联系在一起的规范化的测量。因此在给出量值的同时，必须给出适应于应用目的或实际需要的不确定度或误差范围，否则，所进行的测量的质量就无从判断，量值也就不具备充分的社会实用价值。所谓量值的准确, 即是在一定的不确定度、误差极限或允许误差范围内的准确。一致性是指在计量单位统一的基础上，无论何时、何地，采用何种方法，使用何种计量器具，以及由何人测量，只要符合有关的要求，其测量结果应在给定的区间内一致。也就是说，测量结果应该是可重复、可再现(复现)、可比较的。计量的一致性不仅适应国内，也适应国际，如，国际关键比对和辅助比对的结果，应在等效区间或协议区间内一致。溯源性是指任何一个测量结果或计量标准的量值，都能通过一条具有规定不确定度的连续比较链与计量基准联系起来，使所有的同种量值都可以按照这条比较链通过校准向测量的源头追溯, 也就是溯源到同一计量基准(国家基准或国际基准)，使准确性和一致性得到技术保证。如果量值出于多源或多头，必然会在技术上和管理上造成混乱。假设我国和美国的测量标准没有溯源到同一国际标准，势必造成两国测量结果没有可比性，致使不能实现互认，从而给经济贸易和技术交流等带来障碍。可见量值溯源是测量数据可信性的基础。 “溯源链”应具有规定的测量不确定度。通常要求校准结果的不确定度只能占被校准测量仪器最大允许误差的1/3～1/5，具体取值随测量项目确定。例如，一些无线电参数的测量，能达到1/2，已属不易；而时间频率的测量，往往可达到1/10。 2计量溯源途径

第三章计量溯源性(第一、第二节)-2

第三章计量溯源性主要内容第一节计量溯源性第二节测量标准第三节校准第四节计量检定第五节计量比对第一节计量溯源性一、有关术语及定义 1、计量溯源性 ——通过文件规定的不间断的校准链，测量结果与参照对象联系起来的特性，校准链中的每项校准均会引入测量不确定度。 1）参照对象：可以是实际实现的测量单位的定义，或包括无序量测量单位的测量程序，或测量标准。注：序量——JJF1001-3.28条款：由约定测量程序定义的量。如洛氏硬度HRC标尺:洛氏硬度是以压头压入的深度值计量被测材料硬度值的，规定每压入0.002mm，为洛氏硬度一个单位。HRC是总载荷为150kgf，压头为120o圆锥金刚石测试的硬度值。 2）计量溯源性要求建立校准等级序列。 3）“溯源性”有时是指“计量溯源性”，有时用于其他概念。当有产生混淆的风险时，最好使用全称“计量溯源性”。 2、校准等级序列 ——从参照对象到最终测量系统之间校准的次序，其中每一等级校准

的结果取决于前一等级校准的结果。 1）沿着校准的次序，测量不确定度必然逐级增加。 2）校准等级序列由一台或多台测量标准和按测量程序操作的测量系统组成。 3、计量溯源链 ——用于将测量结果与参照对象联系起来的测量标准和校准的次序。简称溯源链。 1）计量溯源链是通过校准等级关系规定的。 2）计量溯源链用于建立测量结果的计量溯源性。 4、向测量单位的计量溯源性 ——参照对象是实际实现的测量单位定义时的计量溯源性。简称向单位的计量溯源性。向测量单位的计量溯源性是计量溯源性的一种形式。 5、溯源等级图 ——一种代表等级顺序的框图，用以表明测量仪器的计量特性与给定量的测量标准之间的关系。溯源等级图是对给定量或给定类别的测量仪器所用比较链的一种说明，以此作为其溯源性的证据。 6、国家溯源等级图 ——在一个国家内，对给定量的测量仪器有效的一种溯源等级图，包括推荐（或允许）的比较方法或手段。在我国，也称国家计量检定系统表。 7、量值传递 ——通过对测量仪器的校准或检定，将国家测量标准所实现的单位量值通过各等级的测量标准传递到工作测量仪器（工作计量器具）的活动，以保证测量所得的量值准确一致。

CNAS-CL06：2014《测量结果的溯源性要求》

测量结果的溯源性要求Requirements on the Traceability of Measurement Results 中国合格评定国家认可委员会

测量结果的溯源性要求 1．范围本文件规定了CNAS在对检测实验室（含医学实验室）、校准实验室、检查机构、标准物质/标准样品生产者和能力验证提供者等机构（以下统称合格评定机构）实施认可活动时涉及的测量结果的计量溯源性要求。本文件适用于检测（含医学检验）、校准活动，也适用于检查、标准物质/标准样品生产和能力验证等涉及测量活动的合格评定活动。本文件只适用于合格评定机构通过外部机构获得的测量结果的溯源性要求。合格评定机构实施的内部校准应满足CNAS-CL31《内部校准要求》。 2. 规范性引用文件下列文件中对于本文件的应用是必不可少的。凡是注日期的引用文件仅注日期的版本适用于本文件。凡是不注日期的引用文件，其最新版本（包括修改单）适用于本文件。 2.1 CNAS-CL01:2006 检测和校准实验室能力认可准则 2.2 ILAC-P10:2013 ILAC关于测量结果溯源性的政策 2.3 ISO/IEC指南99 国际计量学词汇—基础和通用概念及相关术语（VIM） 3. 术语和定义本文件采用VIM中的有关术语和定义。 3.1计量溯源性（VIM 2.41）通过文件规定的不间断的校准链，将测量结果与参照对象联系起来的测量结果的特性，校准链中的每项校准均会引入测量不确定度。注1：本定义中的参照对象可以是实际实现的测量单位的定义、或包括非序量测量单位的测量程序、或测量标准。注2：计量溯源性要求建立校准等级序列。注3：参照对象的技术规范必须包括在建立校准等级序列时使用该参照对象的时间，以及关于该参照对象的计量信息，如在这个校准等级序列中进行第一次校准的时间。

检验结果量值溯源管理程序修订稿

检验结果量值溯源管理程序 WEIHUA system office room 【WEIHUA 16H-WEIHUA WEIHUA8Q8-

检验结果量值溯源管理程序目的：建立和实施检验结果的量值溯源程序，使病人标本的测量结果能够通过一条具有规定不确定度的连续比较链，或通过实验室间的比对等方式，与测量基准联系起来，从而使测量结果的准确性得到技术保证。 2、范围：适用于检验科开展的检验项目。 3、职责：各专业组长负责本专业组检验项目校准计划的制定以及室间比对计划、自建检测系统校准计划的制定和实施。技术负责人负责校准计划、室间比对计划、自建检测系统校准计划的审批。检验人员负责检验仪器的校准、比对项目标本的准备和检测。 4 定义和术语量值溯源是指测量结果或测量标准的值，能够通过一条具有规定不确定度的连续比较链，与测量基准联系起来，从而使测量结果的准确性和一致性得到技术保证。实验室间比对指按照预先规定的条件，由两个或多个实验室对相同或类似的被测物品进行检测的组织、实施和评价。实验室内部比对

指按照预先规定的条件，由检验科内部实施的同一检验项目不同检测系统之间的比对，也包括各自实验室内部同一检验项目不同检测系统之间的比对。方法学比较实验室准备用一个新的检测系统或测定方法，或新的试剂盒、新的仪器进行病人标本测定前，应与原有的检测系统或公认的参考方法一起检测一批病人标本，以评价新的检测系统或方法引人后的偏倚，从而决定其能否应用于临床；也包括不同检测系统之间所进行的比对。 5 工作程序如果无法实现检验结果的量值溯源或不相关，还可采用以下方法（但不限于此）以提供结果的可信度： a ）对检测系统定期进行校准； b ）参加适当的实验室间比对活动； c ）使用相应的参考物质：此参考物质必须是有资格的供应商提供的有证标准物质，并附有材料特性的详细说明； d ）比率或倒易型测量； e ）使用已明确建立的、经规定的、性能已确定的、被各方承认的协议标准或方法； f ）利用供应商或制造商提供的试剂、程序或检测系统对溯源性的说明，形成实验室的溯源性文件检测系统的校准