No job name

Anal. Chem. 2006, 78, 5900-5905
Lab-on-Valve System Integrating a
Chemiluminescent Entity and In Situ Generation of
Nascent Bromine as Oxidant for Chemiluminescent
Determination of Tetracycline
Mei Yang,†,‡ Ying Xu,‡ and Jian-Hua Wang*,†
Research Center for Analytical Sciences, Northeastern University, Box 332, Shenyang 110004, China, and Department ofChemistry, Liaoning Normal University, Dalian 116029, China A novel configuration of a lab-on-valve (LOV) system was
and FDA have launched the maximum residue limit of 100 µg fabricated and applied for chemiluminescence (CL) detec-
L-1 TCs in milk.4 A number of analytical protocols for TCs were tion by integrating a demountable Z-type flow cell onto the
recently reported,5-11 among which mass spectrometric proce- LOV unit. A bismuthate immobilized microcolumn was
dures showed promising detection capability.5-8 For routine incorporated in one port of the LOV for in situ oxidation
analysis, however, the development of low cost but sensitive and of KBr and generation of bromine as oxidant for the
robust procedures are highly desired.
bromine-hydrogen peroxide-tetracycline (TC) chemilu-
Chemiluminescence (CL) plays an important role and pos- minescent reaction. The nascent bromine reacts with
sesses clear advantages for the analysis of TCs owing to high hydrogen peroxide and produces a weak CL signal, the
sensitivity and simple instrumentation.12-15 Bromine was employed intensity of which was significantly enhanced in the
as an effective oxidant for the CL reaction.16 The sophisticated presence of TC following an energy-transfer mechanism.
procedure for preparing bromine standard solutions in addition A novel procedure for tetracycline quantification was
to its unfavored limit of detection restricted its applications. The therefore developed based on the present system. When
electrogenerated nascent bromine avoids the preparation of compared with the reported flow injection-CL methods for
bromine standard solution;17,18 however, the additional instrumen- TC, this procedure not only provided an improved detec-
tation and the controlling facilities further complicated the entire tion limit of 2.0 µg L-1 but also minimized sample and
system. Furthermore, the stability of the system for generating reagent consumption. A linear range of 6.0-10 000 µg
bromine also needs to be further improved.
L-1 was derived along with RSD values of 5.9 (at the
Flow injection (FI) chemiluminescece procedures facilitate fast concentration level of quantification limit) and 2.2% (at
measurement of a transit CL signal, while their applications in 50 µg L-1), and a sampling frequency of 120 h-1 was
routine analyses are rather limited, attributed in part to the achieved. The system was validated with a National
intricate network of conduits and channels leading to the lack of Standard Procedure (GB/T 18932.4-2002, HPLC with
robustness.19,20 At present, it is a fashion to develop analytical UV detection) by measuring TC contents in commercial
(4) Andersen, W. C.; Roybal, J. E.; Gonzales. S, A.; Turnipseed, S. B.; Pfenning, milk samples.
A. P.; Kuck, L. R. Anal. Chim. Acta 2005, 529, 145-150.
(5) Lindsey, M. E.; Meyer, M.; Thurman, E. M. Anal. Chem. 2001, 73, 4640-
Tetracyclines (TCs), a group of antibiotics, have been em- (6) Hamscher, G.; Sczesny, S.; Hoper, H.; Nau, H. Anal. Chem. 2002, 74, 1509-
ployed for disease prevention and infection treatment and oc- casionally used as additives with low concentrations in stock- (7) Batt, A. L.; Aga, D. S. Anal. Chem. 2005, 77, 2940-2947.
breeding.1 The widespread employment of TCs caused a series (8) Eichhorn, P.; Aga, D. S. Anal. Chem. 2004, 76, 6002-6011.
(9) Goicoechea, H. C.; Olivieri, A. C. Anal. Chem. 1999, 71, 4361-4368.
of problems, as the residues in milk and meat might be either (10) Korpela, M. T.; Kurittu, J. S.; Karvinen, J. T.; Karp, M. T. Anal. Chem. 1998,
directly toxic or cause allergic reactions in some hypersensitive individuals.2 In addition, long-term consumption of foodstuffs with (11) Loetanantawong, B.; Suracheep, C.; Somasundrum, M.; Surareungchai, W.
Anal. Chem. 2004, 76, 2266-2272.
low-level antibiotics poses problems associated with the spread (12) Zhang, X. R.; Baeyens, W. R. G.; Vandenborre, A.; Vanderweken, G.; of drug-resistant microorganisms.3 Therefore, EU, FAO/WHO, Calokerinos, A. C.; Schulman, S. G. Analyst 1995, 120, 463-466.
(13) Halvatzis, S. A.; Timotheoupotamia, M. M.; Calokerinos, A. C. Analyst 1993,
* Corresponding author. E-mail: jianhuajrz@mail.edu.cn. Tel: +86 24 83688944.
(14) Lau, C.; Lu, J. Z.; Kai, M. Anal. Chim. Acta 2004, 503, 235-239.
(15) Townshend, A.; Ruengsitagoon, W.; Thongpoon, C.; Liawruangrath, S. Anal. Chim. Acta 2005, 541, 105-111.
(1) Nozal, L.; Arce1, L.; Simonet, B. M.; Rios, A.; Valcarcel, M. Anal. Chim. (16) Alwarthan, A. A.; Townshend, A. Anal. Chim. Acta 1988, 205, 261-265.
Acta 2004, 517, 89-94.
(17) Zheng, X.-W.; Yang, M.; Zhang, Z.-J. Acta Chim. Sin. 2001, 59, 945-949.
(2) Wangfuengkanagul, N.; Siangproh, W.; Chailapakul, O. Talanta 2004, 64,
(18) Zheng, X.-W.; Yang, M.; Zhang, Z.-J. Anal. Chim. Acta 2001, 440, 143-
(3) Mathur, S.; Singh, R. Int. J. Food Microbiol. 2005, 105, 281-295.
(19) Wang, J.-H.; Hansen, E. H. Anal. Chim. Acta 2002, 456, 283-292.
Analytical Chemistry, Vol. 78, No. 16, August 15, 2006 Figure 1. Schematic diagram of the SI-LOV CL detection system incorporating a bismuthate immobilized microcolumn. Operating procedure:
a small amount of KBr solution was aspirated via port 3 to flow through the oxidation column, where KBr was oxidized to Br2. Sample and H2O2
solutions were then aspirated successively. The stacked zones were afterward dispensed via port 5 toward the flow cell, during which process
the Br -
H2O2 TC CL reaction was facilitated and the CL was monitored in the flow cell.
methodologies for minimized reagent and sample consumption minimized reagent and sample consumption. In addition, the governed by an integrated miniature unit.21,22 At this juncture, the selectivity for tetracycline was improved by employing solid-phase relatively large fluidic consumption by FI does not seem to fit this trend. The recently emerged sequential injection (SI) lab-on-valve (LOV) conception23-28 has proven to be an excellent front EXPERIMENTAL SECTION
end for downscaling the level of fluidic manipulation and sample Apparatus. The lab-on-valve system adopted in the present
pretreatment and thus facilitates novel applications and provides work consists of a few independent components, as illustrated in alternatives for integrating various detection facilities incorporating Figure 1. A Micro-SIA sequential injection system (FIAlab Instru- appropriate entities in the LOV system.
ments Inc., Bellevueo, WA) equipped with a 2.5-mL syringe pump In this study, we report a novel configuration of LOV integrat- was employed for fluidic delivery. The central part of the system, ing a CL detection entity. The latter is designed to be demountable i.e., the LOV unit, integrates all the necessary flow channels and from the main body of the LOV, which facilitates CL detection in sampling ports, in which various reagent-based assays can be a Z-type flow cell. A bismuthate immobilized silica microcolumn performed. A microcolumn packed with sodium bismuthate was incorporated in the LOV for in situ oxidation of KBr, and immobilized silica beads was incorporated into port 3 of the LOV generation of bromine served as oxidant for the bromine- for in situ oxidation of KBr to produce bromine serving as oxidant hydrogen peroxide-TC CL reaction. The nascent bromine reacts for the bromine-hydrogen peroxide-tetracycline CL reaction. A with hydrogen peroxide and produces a weak CL, the intensity Z-type multipurpose flow-through cell, with a channel inner of which was significantly enhanced in the presence of tetracycline, diameter of 1.0 mm and a total channel length of ∼30 mm along following an energy-transfer mechanism.
with a capacity of ∼24 µL, is permanently integrated on the side To the best of our knowledge, this is the first attempt to of the LOV unit and connected to port 5; this part is designed to conduct CL detection in the lab-on-valve and in situ oxidation for be demountable from the main body of the LOV whenever generating nascent bromine by using immobilized bismuthate.
necessary. The flow cell allows a series of miniaturized fluidic When compared with the reported FI-CL procedures, the present operations to be performed, including mixing and dispersion of one provides not only an improved limit of detection, but also fluidic zones. The light emission from the CL reaction system wasmonitored with a side-on photomultiplier tube (PMT, Xi′an Remax (20) Wang, J.-H.; Hansen, E. H.; Miro, M. Anal. Chim. Acta 2003, 499, 139-
Electronics) by directly adhering its light entrance window to the (21) Ruzicka, J.; Hansen, E. H. Anal. Chem. 2000, 72, 212A-217A.
(22) Brivio, M.; Verboom, W.; Reinhoudt. D, N. Lab. Chip 2006, 6, 329-344.
The entire system was controlled with a personal computer (23) Ruzicka, J. Analyst 2000, 125, 1053-1060.
by running the FIAlab for Windows software (FIAlab Instruments) (24) Wang, J.-H.; Hansen, E. H. Anal. Chim. Acta 2001, 435, 331-342
(25) Erxleben, H.; Ruzicka, J. Anal. Chem. 2005, 77, 5124-5128.
and IFFM-D software (Xi′an Remax Electronics).
(26) Long, X.-B.; Miro, M.; Hansen, E. H. Anal. Chem. 2005, 77, 6032-6040.
As compared to the previous versions of lab-on-valve with a (27) Ogata, Y.; Scampavia, L.; Ruzicka, J.; Scott, C. R.; Gelb, M. H.; Turecek, F.
single-channel flow cell integrated into the LOV,29-32 the present Anal. Chem. 2002, 74, 4702-4708.
(28) Long, X. B.; Miro, M.; Hansen, E. H. J. Anal. At. Spectrom. 2005, 20, 1203-
(29) Wang, Y.; Wang, J.-H.; Fang, Z.-L. Anal. Chem. 2005, 77, 5396-5401.
Analytical Chemistry, Vol. 78, No. 16, August 15, 2006 one accommodates a Z-type flow cell on a side part of the main The column was replaced whenever a significant drop of the CL body. This configuration not only is easier to fabricate but offers was recorded. Practically, a much easier way could be used to vast potentials and versatilities in its compatibility with a variety judge whether the column should be renewed. A newly packed of detection modes, i.e., spectrophotometry, laser-induced fluo- column is brown colored, while it gradually turns to light color rescence, electrochemistry, and chemiluminescence.
after a number of operations, and it finally turned to white with All the external channels were 0.8-mm-i.d PTFE tubing con- excessive use, which revealed that most of the immobilized nected to the LOV unit with PEEK nuts/ferrules. The capacity of sodium bismuthate has been consumed, and replacement of the Chemicals. All the reagents used in the present study were
Sample Pretreatment. Tetracycline has the potential to
at least of analytical reagent grade, and 18 MΩ cm-1 deionized chelate metal species and bind onto silanol groups. Thus, for the sample pretreatment process, the employment of a strong chelat- A tetracycline stock solution of 100 mg L-1 was prepared by ing reagent, i.e., EDTA, facilitates the removal of metals and dissolving appropriate amounts of tetracycline hydrochloride improves the recovery of the target species.1,5 Therefore, the (Sigma) in deionized water. Working solutions of various con- centrations were prepared daily by stepwise dilution of the stock A 5.0-mL portion of pH 4.05 Mcllvaine buffer was added to 1.0 mL of the milk sample in a glass centrifuge tube, which was placed A KBr solution of 0.05 mol L-1 was obtained by dissolving 3.0 in the ultrasonic cleaning bath and the mixture was sonicated for g of KBr (Kemiou Chemicals, Tianjin, China) in 0.02 mol L-1 H2- 10 min. After being centrifuged for 20 min at 4000 rpm, the SO4 (Sinopharm Chemical Reagent Co., China, SCRC) and diluted supernatant was collected, and the solid residue was subjected to a second ultrasonic extraction. The total collected supernatant H2O2 working solutions were prepared by appropriate dilution was then filtered through a 0.45-µm PTFE filter, and the filtrate of the 30% (v/v) solution (Kemiou Chemicals) with 0.01 mol L-1 was subjected to solid-phase extraction by adopting a nonpolar NaOH. The use of an alkaline medium deteriorates the long-term silica-based Isolute C18 syringe cartridge (Dalian Institute of stability of the H2O2 solution; therefore, it was prepared daily, and Chemistry and Physics, Chinese Academy of Sciences), with the its optimum concentration is stated in the following.
column sorbent previously activated using an appropriate amount A pH 4.05 Mcllvaine buffer solution was prepared by dissolving of methanol and water. After sample loading, the column was 6.4 g of citric acid monohydrate (SCRC), 14.2 g of Na prewashed with an aliquot of 5.0 mL of Mcllvaine buffer at a flow (SCRC), and 30.25 g of Na2EDTA (SCRC) in 813 mL of deionized rate of 1.5 mL min-1. The retained tetracycline was finally eluted with 2.0 mL of methanol, adopting an elution flow rate of 1.0 mL Other chemicals used were silica gel beads (nominal bead size min-1, the eluate was rotary evaporated to dryness at 40 °C, and of 100 mesh, Wu-si Chemicals Co., Shanghai) and NaBiO3 the residue was dissolved in 500 µL of water for the ensuing CL Deionized water was used throughout as carrier solution.
Operating Procedure. An entire operating cycle includes the
Bismuthate Immobilization and Preparation of the in Situ
following steps: 800 µL of carrier was aspirated into the syringe, Column Reactor. A 5.0-g sample of silica beads was treated in
which was afterward used to deliver the sample and reagent zones an oven at 1000 °C for 3 h and afterward mixed with 1.0 g of toward the flow cell. Subsequently, 15 µL of KBr solution was NaBiO3 fine powder. The mixture was stirred for 30 min, during aspirated via port 3 of the LOV at a flow rate of 20 µL s-1. As this which process the NaBiO3 was immobilized onto the surface of zone flowed through the NaBiO3 immobilized column, the KBr the silica beads. A microcolumn was subsequently packed by was oxidized giving rise to nascent bromine, which was stored in using ∼100 mg of the obtained silica beads in a piece of PTFE the holding coil. An aliquot of 50 µL of sample solution was then tubing (2.0-mm i.d., 3.2-mm o.d.) blocked at both ends by glass aspirated into the holding coil via port 4 followed by 50 µL of H2O2 wool. The microcolumn was then integrated into port 3 of the solution via port 2, both at a flow rate of 50 µL s-1. Thereafter, lab-on-valve unit. Before use, the column was rinsed thoroughly the stacked zones were dispensed via port 5 toward the flow cell, by pumping 15 mL of H2SO4 (0.02 mol L-1) and 15 mL of deionized during which process the bromine-hydrogen peroxide-tetracy- water alternatively through it at a flow rate of 1.5 mL min-1, to cline CL reaction was facilitated and the light emission was ensure the complete removal of the loosely adhered NaBiO3 on monitored in the flow cell using a side-on PMT operated at a negative high voltage of 900 V. The blank level of the detection Further experiments demonstrated that no obvious deteriora- system was occasionally checked and adjusted by introducing an tion of the oxidizing capability of the column was observed after equal amount of carrier (50 µL) instead of sample solution into 300-400 cycles for complete conversion of 15 µL of KBr solution (0.05 mol L-1) to bromine. The performance of the column was The NaBiO3 immobilized microcolumn was replaced whenever checked, alternatively monitoring the emitted CL intensity by it was excessively used, i.e., after ∼350 samples have been adopting a tetracycline standard solution of affixed concentration.
processed using the present system. A rinsing procedure wasadopted for the new column, as described in the section Bismuth- (30) Chen, X.-W.; Wang, W.-X.; Wang, J.-H. Analyst 2005, 130, 1240-1244.
(31) Hindson, B. J.; Brown, S. B.; Marshall, G. D.; McBride, M. T.; Makarewicz.,
A. J.; Gutierrez, D. M.; Wolcott, D. K.; Metz, T. R.; Madabhushi, R. S.;
Dzenitis, J. M.; Colston, B. W. Anal. Chem. 2004, 76, 3492-3497.
RESULTS AND DISCUSSION
(32) Edwards, K. A.; Baeumner, A. J. Anal. Chem. 2006, 78, 1958-1966.
Flow Cell Configurations. In the present chemiluminescent
(33) Cinquina, A. L.; Longo, F.; Anastasi, G.; Giannetti, L.; Cozzani, R. J. Chromatogr., A 2003, 987, 227-233.
system, the configuration of the flow cell has been proven to affect Analytical Chemistry, Vol. 78, No. 16, August 15, 2006 significantly the recorded CL signal. The experiments indicatedthat, when using a conventional spiral flow cell with channel innerdiameter of 0.5 cm instead of the Z-type flow cell, a thoroughmixing of the stacked zones occurred as they flow from a narrowchannel in the LOV into a wider one. Thus, a 15-20-fold higherblank signal was recorded attributed to the reaction of nascentbromine and hydrogen peroxide. Consequently, a deteriorationof the signal/noise ratio was observed. It is conceivable that theblank signal should be lower if a spiral flow cell of smaller diameterwere used. Actually, this was well demonstrated by the Z-typeconfiguration of the flow cell. The present Z-type flow cell wasdesigned to fit the receiving area of the light entrance window ofthe PMT, which provides sufficient CL detection area by incor-porating the Z-channel onto an appropriate planar area and, Figure 2. Effects of sampling sequences on the recorded CL.
therefore, facilitates a complete collection of the light emission.
Conditions: KBr concentration in 0.02 mol L-1 H2SO4, 0.05 mol L-1; As compared with the single-purposed spiral cell for CL detection, H2O2 concentration in 0.01 mol L-1 NaOH:, 0.01 mol L-1; sampling the multipurposes of the Z flow cell are characterized by the vast (aspirating) sequence, Br2 sample-H2O2; volumes of KBr-sample-H2O2 solutions (µL), 15-50-50; aspirating flow rates for KBr- potentials and versatilities in its compatibility with a variety of sample-H2O2 solutions (µL s-1), 20-50-50; negative high voltage detection modes, i.e., spectrophotometry, laser-induced fluores- cence, electrochemistry, and chemiluminescence. In addition,Z-type channels are much easier to be fabricated mechanically,while the fabrication of spiral channels into the monolithic LOV silica surface; there is thus a risk of losing particulate NaBiO3 system poses a real challenge, unless microfabrication techniques during the oxidation process. As a compromise, a mass ratio of 1:5 was employed for the immobilization process, yielding both a In Situ Generation of Nascent Bromine and the Perfor-
satisfactory oxidizing efficiency and an adequate lifetime of 300- mance of the Immobilized Bismuthate Column. The overall
performance of the entire system depends strongly on the NaBiO3 has very strong oxidizing ability in acidic medium. To concentration of the nascent bromine. Therefore, a high oxidation maintain the oxidizing capability and the stability of the packed column, an appropriate acid should be employed. The experiments oxidation rate is highly desired. It is ideal that quantitative indicated that HCl and HNO3 destruct the packed column and oxidation is accomplished while the KBr solution was aspirated therefore were not a choice. When using H3PO4 and H2SO4, a to flow through the column reactor. The oxidation kinetics was higher response and a longer lifetime of the column were therefore investigated by varying the contact time of KBr solution achieved. For the ensuing experiments, H2SO4 was employed.
(15 µL) with the oxidizing surface, which was realized by changing Further investigations showed that an improvement on the the flow rate for aspirating the KBr solution through the column.
recorded CL intensity was observed with the increase of H2SO4 A complete oxidation of KBr in a 0.05-0.30 mol L-1 solution was concentration in KBr solution up to 0.02 mol L-1, while afterward, achieved within 1 s, by aspirating 15 µL of solution to flow through a decline of the signal was recorded. This might be attributed to the column reactor at a flow rate of 5-25 µL s-1. When employing the fact that the present CL reaction is preferentially performed KBr solution within this concentration range and the optimized in alkaline medium, while a higher concentration of acid tends to experimental conditions as detailed in the ensuing sections, a dilute the concentration of NaOH in the adjacent H2O2 zone, which virtually constant CL intensity was recorded. This result denotes in turn deteriorates the CL reaction. For further experiments, a that a KBr concentration of 0.05 mol L-1 with an aspirating flow H2SO4 concentration of 0.02 mol L-1 was employed.
rate of 20 µL s-1 is adequate for the ensuing CL reaction for the Effect of Sampling Sequence. For the present SI-LOV CL
reaction system, the penetration and dispersion of the stacked When using a fixed amount of NaBiO3 immobilized silica for zones, i.e., Br2 in H2SO4, H2O2 in NaOH, and tetracycline (sample), packing the column reactor, no variations for the oxidation is the decisive factor on the CL emission. Therefore, the effect of efficiency were observed by varying the length and diameter of sampling sequence of the various zones and their concentrations the packed column. Thus, for the ensuing experiments, ∼100 mg and volumes were carefully optimized.
of NaBiO3 immobilized silica beads was employed to pack the Figure 2 illustrated the effect of sampling sequence of the three column into a PTFE tubing of 2.0-mm inner diameter.
zones. It is obvious that a very weak emission was detected The immobilized amount of NaBiO3, which is directly related whenever the Br2 and H2O2 zones were directly adjacent, while to the amount of NaBiO3 used for immobilization, has a significant significant emissions were recorded in the cases where the Br2 effect on the oxidizing efficiency and the lifetime of the column.
and H2O2 zones were separated by the sample zone. It is When a NaBiO3/silica mass ratio of 1:10 (m/m) was adopted, the interesting that the highest CL intensity was always achieved by lifetime of the packed column is only ∼100 operating cycles for adopting the sampling (aspirating) sequence of Br - 15 µL of KBr solution. Although the oxidizing efficiency and sample-Br2. This observation could be satis- lifetime of the column were significantly improved with a mass factorily explained by the following energy-transfer mechanism ratio of 1:1, the NaBiO3 was not uniformly immobilized on the Analytical Chemistry, Vol. 78, No. 16, August 15, 2006 When adopting the sampling sequence of Br - sample-Br2, the Br2 and H2O2 zones penetrate through the sample zone and contact each other while flowing toward theflow cell. At the very moment when the nascent Br2 meets H2O2,the excited state of O * 2 is generated, which immediately transfers its energy to the TC molecule and produces the excited state of Figure 3. Effects of various zone volumes on the recorded CL
TC*. Afterward, the transition of TC* to ground state along with signal. Conditions: KBr concentration in 0.02 mol L-1 H2SO4, 0.05 the emission of strong CL takes place. When the zones are mol L-1; H2O2 concentration in 0.01 mol L-1 NaOH, 0.01 mol L-1; directed toward the flow cell, an alkaline medium is preferential to the CL reaction. At this point, the leading part of the flow for rates for KBr-sample-H2O2 solutions (µL s-1), 20-50-50; negativehigh voltage of the PMT, 900 V.
medium for the CL reaction and, thus, gives rise to strongemission. In the case of H By taking into account the fast kinetics of the present reaction part of the flow is acidic. When flowing toward the flow cell, an and the transit nature of the CL signal, the recorded readout acidic wetting film was formed on the inner wall of the channel, depends strongly on the flow rate of the various zones. Thus, a which slightly diluted the alkaline medium of the ensuing H2O2 faster dispensing flow rate of 120 µL s-1 was employed in order zone and, therefore, a decline of the recorded CL.
to minimize the time required for transporting the stacked zones For the sampling sequences where Br2 and H2O2 zones were (reaction mixture) into the detection point in the flow cell. An directly adjacent, although a sufficient amount of excited state of even higher flow rate, i.e., 150 µL s-1, resulted in virtually no was generated, there is no chance to transfer its energy to increment of the recorded signal but a slight deterioration on the majority of its excitation energy was released in the form of heat.
Effect of Foreign Species. For real world samples, the matrix
Therefore, only very weak signals were detected.
compositions are usually quite complex, which might cause Sample and Reagent Zone Volumes and Flow Rates. Based
interferences for the determination of trace-level TCs. Thus, on the discussions in the above sections, the stacked zone volumes potential interfering effects from some of those species frequently and their concentrations should be carefully chosen to achieve encountered in analyzing livestock products, i.e., milk and meat, optimized CL. Figure 3 illustrated the effects of zone volumes on were investigated by using the present system.
the recorded CL intensity. The selection of Br2 zone volume is For the assay of 1.0 mg L-1 tetracycline, no interfering effects quite tricky; the amount of nascent bromine generated from a were observed in the presence of 1000-fold of alkali metal ions, small zone of 10 µL is not sufficient for the CL reaction, while a large zone corresponds to a large amount of H2SO4, which tends earth and the majority of heavy metal species, and 10-fold of to deteriorate the alkaline medium for the reaction. Thus, a Br2 ascorbic acid and uric acid. However, equal amounts of Cu2+, Zn2+, zone of 15 µL was employed throughout the study.
cations cause interfering effects. In some particular The H2O2 zone provides an alkaline medium for the CL cases, appropriate dilution of the sample solution or digestion reaction. The experiments have shown that the CL intensity helps to attenuate the matrix effects, but this cannot be used reached maximum by using a H2O2 zone volume of 50-100 µL; a excessively because TC contents in many real world samples were further increase of the zone volume results in an enhancement quite low. It is therefore highly desired to incorporate a sample of blank and thus a decline of the net signal, which is a common pretreatment procedure in order to eliminate the interfering observation in CL reactions involving H2O2. Further experiments components prior to analysis. In the present case, when employing showed that a maximum CL intensity was recorded when a solid-phase extraction procedure with a C18 cartridge as detailed employing a H2O2 concentration of 0.01 mol L-1 in a 0.01 mol L-1 in the Sample Pretreatment section, the potential interfering NaOH solution. For further studies, a 50-µL zone volume of this species mentioned herein could be effectively eliminated.
Analytical Performance of the Present System and Its
It is conceivable that an appropriate sample zone not only Validation. The characteristic performance data of the SI-LOV
facilitates the penetration of Br2 and H2O2 zones in order to trigger CL detection system for TC were summarized in Table 1, along their reaction and give rise to the excited state of O * with a comparison to some of the reported flow injection CL able to receive the energy transferred from the O * procedures, in terms of limit of detection and reagent/sample a small sample zone is not sufficient to receive the transferred consumption. It is obvious that the present system not only energy, while too large a sample zone makes it difficult to be significantly improved the limit of detection for the CL reaction effectively penetrated/dispersed through by the Br2 and H2O2 system with bromine as the oxidant but also substantially reduced zones. This requirement was only fulfilled when using a sample the consumption of reagent. In addition, the present system zone volume of 50-150 µL and a steady CL signal was recorded.
provides a sampling frequency of 120 h-1, which is a clear Therefore, a sample zone of 50 µL was used.
advantage for routine analysis. Furthermore, the high precision Analytical Chemistry, Vol. 78, No. 16, August 15, 2006 Table 1. Characteristic Performance Data for the SI-LOV CL Detection System with Nascent Bromine as Oxidant
for the Determination of Tetracycline
I ) 0.5451CTC (µg L-1) + 49.35 RSD (at the level of quantification limit) Features of FI-CL Procedures for Tetracycline Table 2. Determination Results for Tetracycline in Milk
attributed to the fact that the original concentrations achieved by Samples by a National Standard Procedure (GB/T
18932.4-2002) and the Present System
the standard procedure were roughly 10% higher, and this errorwas transferred to the final recoveries.
CONCLUSIONS
By integrating a Z-type flow cell serving as the CL detection entity, a novel SI-LOV system was employed for CL detection.
The incorporation of a bismuthate immobilized microcolumn onto the LOV gives rise to nascent bromine by in situ oxidation of KBr, LOV system not only significantly improved the limit of detection and stability along with a long lifetime of the packed column for tetracycline but also minimized sample and reagent consump-tion. In addition, the simple instrumentation makes it possible to reactor guarantee long-term stability of the entire system, char- construct a portable analyzer for field measurement of antibiotics, acterized by a reasonable RSD value at a very low concentration in addition to the feature of easy automation of the entire system.
Further investigations are under progress. Another distinct feature The present SI-LOV CL detection system with nascent bromine of the present version of lab-on-valve is characterized by the vast as oxidant for tetracycline assay was validated using a National potentials and versatilities in its compatibility with a variety of Standard Procedure for tetracycline (GB/T 18932.4-2002, HPLC detection modes, including spectrophotometry, laser-induced with UV detection issued by the General Administration of Quality fluorescence, electrochemistry, and chemiluminescence.
Supervision Inspection and Quarantine, China) by measuring ACKNOWLEDGMENT
tetracycline contents in commercial milk samples. The sample The authors are indebted to the financial support from the pretreatment was carried out following the procedure detailed in National Natural Science Foundation of China (NSFC-20375007, the Sample Pretreatment section. The results were summarized 20575010, NSFC-RFBR program-20411120643), the key project forscientific research from the Ministry of Education, China (105056), in Table 2. Agreements were achieved between the results the Natural Science Foundation of Liaoning Province, China obtained by the standard procedure and the present system. In (20042011), and the SRFDP program (20050145026).
addition, reasonable spiking recoveries for the commercial milk samples were also attained by employing the present procedure.
Received for review April 19, 2006. Accepted May 23, It can be noted that the recovery rates obtained by using the standard procedure were overall on the low side, which might be Analytical Chemistry, Vol. 78, No. 16, August 15, 2006

Source: http://hxsy.lnnu.edu.cn/Public/lw/91/1.pdf