Anal. Chem. 2002, 74, 5564-5568
Fluorescence Assay Based on Preconcentration by a Self-Ordered Ring Using Berberine as a Model Analyte Ying Liu,†,‡ Cheng Zhi Huang,*,† and Yuan Fang Li†
Institute of Environmental Chemistry, College of Chemistry and Chemical Engineering, Southwest Normal University, Chongqing 400715, China, and Department of Chemistry, Inner Mongolia Normal University, Huhhot 010022, the People’s Republic of China
A novel assay for trace amounts of fluorescent analytes is proposed based on the assembly of a self-ordered ring (SOR) through capillary flow in a sessile droplet on a glass slide support. After solvent evaporation of the sessile droplet containing a fluorescent analyte on a hydrophobictreated glass slide, an outward capillary flow of the solvent from the interior of the droplet occurs. The resultant outward capillary flow then carries the analyte to the perimeter of the droplet spot where the analyte deposits and forms a fluorescent SOR. For the model analyte of berberine, SORs with outer diameter less than 1.2 mm and ring belt width less than 19 µm can be obtained depending on the droplet volume of the berberine solution. Data analysis for the digitally imaged SOR by using a CCD camera showed that the berberine molecules across the SOR belt section follow a Gaussian distribution, and the maximum fluorescent intensity (Imax) was found to be proportional to berberine content at the femtomole level. With the proposed technique, the content in tablets and the average excretion rates of berberine through human urine after oral administration could be satisfactorily monitored.
since the spot takes a shape ranging from solid round9-14 to square.15 The evaporation of a sessile droplet on a substrate is a wellknown phenomenon observed from a spilled drop of coffee drying on a solid surface to the annoying ringlike spots left on dishes. When a droplet of dilute solution is spotted onto a hydrophobic surface, because of the solvent evaporation from the edge wedge of the drop spot, which must be replenished by an outward flow of the solvent from the interior, almost all low molecular weight solutes dispersed in a drying drop are dragged to the perimeter of the spot by this flow, where they accumulate to form the ringlike deposit, commonly known as “the coffee-spot effect”, or according to Deegan, the capillary flow effect.13 Such ring formation in an evaporating sessile drop on the substrate is very common, and it not only influences processes such as printing, washing, and coating12 but also has been used to develop high-throughput automatic DNA mapping and created arrays of DNA spotted for gene expression analysis on the solid substrate.16-20 We expect that the formed self-ordered ring (SOR), when coupled with a fluorescent microscopic image technique, can be used for sensitive determination of trace amounts of fluorescent drugs, medicines,
The dried spot technique has proved a promising analytical tool since it has many advantages, allowing the technique to approach the detection capabilities of inductively coupled plasma mass spectrometry1 and digitally imaging facilities such as the CCD camera.2 It involves the transfer of a small amount of a liquid sample onto a solid support, on which the liquid sample then undergoes drying and leaves a small solid residue, and then the dried spot produces a specimen on the surface of the solid support. Thus, the analytes are concentrated prior to analysis, and the matrix effect resulting from different solutions is strongly reduced.1 However, the challenge for sensitivity improvement for spot analysis exists in the contribution of the solid support such as a thin-film substrate including octadecylsilanized silica and poly(vinyl alcohol) plate.3,4 In addition, analytical signals generally were taken from the whole spot even if in microchip array systems 5-8
(3) Kaneko, E.; Yoshimoto, K.; Yotsuyanagi, T. Chem. Lett. 1999, (8), 751752. (4) Ishidal, A.; Kaneko, E.; Yotsuyanagi, T. Chem. Lett. 1999, (3), 217-218. (5) Taton, T. A.; Mirkin, C. A.; Letsinger, R. L. Science 2000, 289, 1757-1760. (6) Taon, T. A.; Lu, G.; Mirkin, C. A. J. Am. Chem. Soc. 2001, 123, 5164-5165. (7) Sun, Y.; Walker, G. C. J. Phys. Chem. B 2002, 106, 2217-2223. (8) Barkko, A. P.; Dickson, R. M. J. Phys. Chem. B 1999, 103, 11237-11242. (9) Maenosono, S.; Dushkin, C. D.; Saita, S.; Yamaguchi, Y. Langmuir 1999, 15, 957-965. (10) Annarelli, C. C.; Fornazero, J.; Bert, J.; Colombani, J. J. Eur. Phys. E 2001, 5, 599-603. (11) Salamanca, J. M.; Ciampi, E.; Faux, D. A.; Glover, P. M.; McDonald, P. J.; Routh, A. F.; Peters, A. C. I. A.; Satguru, R.; Keddie, J. L. Langmuir 2001, 17, 3202-3207. (12) Deegan R. D.; Bakajin O.; Dupont T. F. Nature 1997, 389, 827-829. (13) Deegan R. D. Phys. Rev. E 2000, 61, 475-485. (14) Deegan R. D.; Bakajin, O.; Doupont T. F. Phys. Rev. E 2000, 62, 756-765. (15) Sapsford, K. E.; Liron, Z.; Shubin, Y. S.; Ligler, F. S. Anal. Chem. 2001, 73, 5518-5524. (16) Abramchuk, S. S.; Khokhlov, A. R.; Iwataki, T.; Oana, H.; Yoshikawa, K. Europhys. Lett. 2001, 55, 294-300. (17) Hu, H.; Larson, R. G. J. Phys. Chem. B 2002, 106, 1334-1344. (18) Murray, C. B.; Kagan, C. R.; Bawendi, M. G. Science 1995, 270, 13351338. (19) Maenosono, S.; Dushkin, C. D.; Saita, S.; Yamaguchi, Y. Langmuir 1999, 15, 957-965. (20) Weiss, P. S.; Yokota, H.; Aebersold, R. J. Phys. Condens. Matter 1998, 10, 7703-7712.
* To whom all correspondence should be addressed. E-mail: chengzhi@ swnu.edu.cn. † Southwest Normal University. ‡ Inner Mongolia Normal University. (1) Link, D. D.; Kingston, H. M., Havrilla, G. J.; Colletti, L. P. Anal. Chem. 2002, 74, 1165-1170. (2) Blossey, R.; Bosio, A. Langmuir 2002, 18, 2952-2954.
5564 Analytical Chemistry, Vol. 74, No. 21, November 1, 2002
10.1021/ac0259358 CCC: $22.00
© 2002 American Chemical Society Published on Web 10/03/2002
and other fluorescent analytes since it is based on signal detection of a formed ring less than 1.2 mm in outer diameter, greatly different from the signal detection of the whole spot ranging from round to square in shape.5-15 Since berberine is an important antiinflammatory drug for heart and intestinal disorders,21,22 is antitumor promotion active and antilipase effective,23 its quantification in tablets is required, and the pharmacologic investigation is compulsory,24-27 we thus take berberine as a model analyte and develop a novel method based on the formed SOR to quantify its content in tablets and monitor the average excretion rates of berberine through human urine after oral administration. EXPERIMENTAL SECTION Apparatus. The SOR was observed under an IX-70 inverted microscope system, equipped with a 100-W mercury arc lamp and a U mirror cube unit with excitation filter of 330-385 nm, barrier filer of 400-800 nm, and dichroic mirror of DM 400 nm (Olympus, Tokyo, Japan). The SOR image was captured by employing a Cohu 4910 series cooled CCD (Cohu, CA) coupled with Scion Image software package for Windows. An MVS-1 vortex mixer (Beide Scientifc Instrumental Ltd., Beijing, China) was used to blend the solutions in 0.5-mL microtubes. An S-10A digital pH meter (Xiaoshan Scientific Instruments Plant, Zhejiang, China) was used to measure the pH values of the aqueous solution. The Origin 5.0 software package was used for the linear and Gaussian fit. Reagents. A 1.00 × 10-3 M stock solution of the hydrochloride salt of berberine (E. Merck, Darmstadt, Germany) was prepared by directly dissolving the commercially purchased product into doubly distilled water. The working solution was 1.00 × 10-5 M. A 0.6% (v/v) solution of poly(vinyl alcohol)-124 (PVA-124, Shanghai Chemical Reagent Co., Shanghai, China) in water and a 4.5% (v/v) solution of dimethyldichlorosilane (DMCS, The First Chemical Reagent Plant of Shanghai, Shanghai, China) in toluene were used. HAc-NaAc buffer solution (pH 3.95) was used to control the acidity. All reagents were of analytical-reagent grade without further purification. Doubly distilled water was used throughout. Pretreatment of Glass Solids. Glass slides were cleaned and pretreated mainly according to ref 28. To clear off the oil impurities of the surfaces, they were first ultrasonically washed with detergent and then 18 M H2SO4 solution saturated with K2Cr2O7, respectively. After being washed with doubly distilled water and acetone, the glass slides were dried by nitrogen gas flow, and immersed in 4.5% (v/v) DMCS in toluene at room temperature. The residue of the DMCS-toluene was removed at last by (21) Zeng, X, J.; Zeng, X. H. Biomed. Chromatogr. 1999, 13, 442-444. (22) Sakai, T. Analyst 1983, 108, 608-614. (23) Zhang, X. B.; Li, Z. Z.; Guo, C. C.; Chen, S. H.; Shen, G. L.; Yu, R. Q. Anal. Chim. Acta 2001, 439, 65-71. (24) Jia, Y. Y.; Zhao, B. N.; Gou, J. G. Chin. Traditional Pat. Med. 1995,17, 17-18. (25) Sakai, T.; Ohno, N.; Chung, Y. S.; Nishikawa, H. Anal. Chim. Acta 1995, 308, 329-333. (26) Shen, G. L.; Yao, S. Z.; Jiang, X. H. J. Chin. Anal. Chem. 1983, 11, 481484. (27) Ji, S. G.; Chai, Y. F.; Zhang, G. D.; Wu, Y. T.; Liang, D. S.; Xu, Z. M. Biomed. Chromatogr. 1999, 13, 439-441. (28) Hergenrother, P. J.; Depew, K. M.; Schreiber, S. L. J. Am. Chem. Soc. 2000, 122, 7849-7850. (29) Zheng, X. Y.; Peng, Y.; Ren, D. Q.; et al. The Pharmacopoeia of the People’s Republic of China; Chemistry and Chemical Engineering Press: Peking, 2000; 2nd Section, p 548-550.
Figure 1. A SOR of berberine formed on a DMCS pretreated glass slide (A) and its fluorescence features across the center section (B). The outer diameter (2R) is 1.2 mm, and the SOR belt width (2δ) is 19 µm. A 4× objective was employed. Spotted solution: berberine, 1.20 × 10-6 M; PVA, 0.06%; pH, 3.95. Droplet volume, 0.50 µL.
immersing in CHCl3 and in acetone, respectively, and then the glass slides were dried with nitrogen gas flow. Pretreatment of Samples. Tablets of berberine were purchased from Kangfulai Pharmaceutical Ltd. (Chengdu, China) and Taiji Pharmaceutical Ltd. (Chongqing, China), respectively. Pretreatment was done according to ref 29. Briefly, the sugar coat was peeled off first, and the residue of the tablets was then pounded into powders. About 0.18 g of the fine powder was directly dissolved in boiling water to prepare a 250-mL sample solution. Human urine samples were collected at different time intervals from three healthy volunteers of the Southwest Normal University Hospital after they took a single 120-mg oral dose. After being centrifuged for 10 min at 500 rpm, the supernatant fluid of the urine samples was collected and 25-fold diluted with doubly distilled water. General Procedure. Into a 0.5-mL microtube were added the appropriate working solution of berberine or sample and 100 µL of HAc-NaAc buffer solution. After being vortex-mixed, 50 µL of 0.60% PVA-124 solution was added and the mixture was diluted with doubly distilled water to 0.5 mL and mixed thoroughly. Then, 0.10-1.00 µL of the mixture was spotted on the surfaces of the pretreated glass slides. The glass slides were immediately transferred to a 70 °C oven for 5-7 min; the formed SOR image was then observed and measured under the Olympus IX 70 inverted microscope system. RESULTS AND DISCUSSION Features of SORs. Berberine is fluorescent under the excitation of ultraviolet light. Figure 1A shows a typical fluorescent SOR image of berberine formed on a DMCS pretreated glass slide. Analytical Chemistry, Vol. 74, No. 21, November 1, 2002
5565
Figure 2. Gaussian fit of peaks a and b in Figure 1. Spline line with square symbol represents original data; straight line without symbol displays Gaussian fit. Peak a: I ) 4.57 + 128.4/(π)1/2e-0.15(x-14.59)2, (χa2 ) 1.191, σa ) 1.802 pixels, ∼8.60 µm). Peak b: I ) 5.99 + 126.0/ (π)1/2e-0.16(x-254.8)2, (χb2 ) 1.900, σb ) 1.765 pixels, ∼8.42 µm).
Because of the transport of berberine solute dragged by the outward capillary flow of solvent (water) in the evaporating sessile droplet on the glass slide, we can see that almost all bererine molecules have been deposited along the perimeter of the spotted droplet. The outer diameter (2R) of the formed SOR was ∼1.2 mm if 0.5 µL of berberine solution was spotted on the glass slide, and the SOR belt width (2δ) was 19 µm. Figure 1B displays the distribution of berberine molecules across the SOR center. It can be seen that the SOR is symmetrical and the fluorescence intensity in and out of the SOR belt is near to zero. It is obvious that the contribution of background is very small, which is a big challenge for sensitivity improvement for spot analysis on a solid support such as a thin-film substrate including octadecylsilanized silica and poly(vinyl alcohol) plate.3,4 By fitting the fluorescence intensity data of Figure 1B, we found that the distribution of berberine molecules across the SOR belt section follows a Gaussian function (Figure 2, χa2 ) 1.191, χb2 ) 1.900), and the σ-values, halfbandwidth of the Gaussian curve where the fluorescence intensity is 0.607Imax (Imax, florescence intensity located at the center of the SOR belt), are much close to each other (Figure 2, σa ) 1.802 pixels, σb ) 1.765 pixels). In our experiments, 1 pixel is ∼4.8 µm when a 4× object was used). This Gaussian distribution of the molecules on the SOR is identical to the reports of other authors.30 This observable Gaussian distribution, according to Blossey,2 originates from the solute diffusion process on microscopic scale and can be geometrically understood as these solutes, which at first are pushed into the collapsing wedge of the droplet, block the later ones from entering the farthest part of the wedge. Thus, the deposition process depends on the initial concentration of the solution and the profile of the shrinking droplet. If solute concentration is too low, the ring structures appear to be poorly developed due to fluctuation effects of solvent evaporation; if the solute concentration is too high, the system reaches a regime dominated by viscous effects of the solute hindering ring formation.30 As Figure 3 shows, the SOR belt width (2δ) will slightly increase with increasing initial berberine concentration under the (30) Latterini, L.; Blossey, B.; Hofkens, J.; Vanoppen, P.; De Schryver, F. C.; Rowan, A. E.; Nolte, J. M. Langmuir 1999, 15, 3582-3588.
5566
Analytical Chemistry, Vol. 74, No. 21, November 1, 2002
Figure 3. Dependence of SOR belt width (2δ) on the concentration of berberine solution. Spotted solution: PVA-124, 0.06%; pH, 3.95. A 4× objective was used. Droplet volume (µL): A, 1.0; B, 0.50; C, 0.30.
same droplet volume, but the outer diameter of the ring is changeless. That is easy to understand according to the blocking mechanism since the blocking effect of earlier deposited molecules acts on the later ones. To the contrary, both the ring belt width (2δ) and the size of the ring (2R) will increase with increasing droplet volume. Dependence of the Size of SORs. It is easy to assume that the SOR size is strongly dependent on the droplet volume. If one imagines that a small aqueous droplet (0.x µL) is spotted onto a hydrophobic substrate, the spotted drop can be regarded as a segment before the solvent evaporation occurs; thus, the contact angle of the spotted droplet with substrate, θ, can be expressed as31
tg(θ/2) ) h/R
(1)
where h is the height of the segment and R is the radius of the spotted drop on the substrate. Due to the hydrophobic features (31) Zhao, G. X. Physical Chemistry of Surfactants; Peking University Press: Beijing, 1984; p 347.
of the DMCS-pretreated surfaces, the contact line is pinned to its initial position and the R-values will not change with the evaporation of the solvent.13 So, R is in fact the outer radius of the SOR. Hence, the volume of the segment, V, can be expressed as32
V ) (1/6)πh(3R2 + h2)
(2)
By substituting eq 1 into eq 2 and rearranging, we can get
R ) KV1/3
(3)
where K ) {6/πtg(θ/2)[3 + tg2(θ/2)]}1/3 and is strongly dependent on the contact angle of the spotted drop with substrate, θ, which is decided by the hydrophobic features of the substrate that were pretreated with DMCS. Equation 3 shows that the outer radius of the SOR is in proportion to the cube root of the droplet volume. We tested the relationship between radius of SOR and droplet volume by using three different glass slides treated in three batches, respectively. It was found that eq 3 was perfectly obeyed in the range of 0.10-3.00 mL of spotted solution with linear regression equations of R ) -1.357 + 156.1V1/3 (r ) 0.9994, n ) 12), R ) 1.263 + 158.1V1/3 (r ) 0.9993, n ) 12), and R ) 1.074 + 163.2V1/3 (r ) 0.9994, n ) 12). In addition, by using different pieces of glass slides, reproducible results could be available for the same volume droplet, indicating the pretreatment procedures for the glass slides are reasonable. Optimization of the General Procedure. It was found that the fluorescence features of SOR depend on the pH values of the spotted solution. After tested buffer solutions, including HAcNaAc, Britton-Robinson, Tris-HCl, hexahydropyridine hydrochloride, and hydropyridine hydrochloride, we found that the Imax value is the strongest only when HAc-NaAc buffer is used to control the pH values of the spotted solution. Optimal pH values were found in the range of 3.81-4.20, and any pH values out of this range lower the Imax values. In this assay, we kept the aqueous medium at pH 3.95. It has proved that PVA is the most effective ring-forming assistant among several kinds of water-soluble polymers, but its concentration is crucial to the SOR formation. If the concentration of PVA were too low, the contact line of the drying sessile droplet will not be pinned, and the ring-shaped deposit is not formed on the hydrophobic substrate. On the contrary, reduced Imax values were observed if high PVA concentrations were employed. This observation was always accompanied with an increase of σ-values of the Gaussian curves. The possible reason is due to the viscosity increase of the solution with increasing PVA concentration, since high viscosity can also modify the deposition of analyte by preventing the droplet from attaining an equilibrated shape.33,34 More seriously, some part of the analyte may be included in the highly viscous PVA membrane, which keeps the solute from transporting along the outward capillary flow of the solvent.12 For this assay, the appropriate concentration of PVA-124 is 0.06%. In addition, the evaporation velocity of the solvent in an oven can be controlled by temperature at the humidity in our laboratory, (32) Zhu, D. X. Elementary Geometry Research; Higher Education Press: Beijing, 1998; p 268. (33) Liu, Y. Z.; Zhang, Z. G. Encyclopedia of Chemical Engineering; Chemical Engineering Press: Beijing, 1995; Vol. 9, p 517. (34) Fan, M. K.; Huang, C. Z.; Li, Y. F. Anal. Chem. Acta 2002, 453, 97-104.
Table 1. Analytical Parameters of SOR Methoda droplet linear range vol (µL) (×fmol/ring) 0.10 0.20 0.30 0.50 1.00
6.9-160.0 8.9-320.0 70.1-570.0 60.8-950.0 52.5-1900.0
corr coeff LOD (3σ, (r, n ) 7) fmol/ring)
linear regressn eq (m, fmol/ring) ∆I ) -0.94 + 0.57m ∆I ) -3.46 + 0.44m ∆I ) -0.14 + 0.16m ∆I ) -0.81 + 0.13m ∆I ) 0.10 + 0.80m
0.9988 0.9988 0.9995 0.9999 0.9997
0.69 0.89 7.01 6.08 5.25
a Spotted solution: PVA-124, 0.06%; pH, 3.95. A 10× objective was used for observing SOR by spotting 0.10- and 0.20-µL solutions. A 4× objective was used for other SORs. The above data can also be expressed as the berberine concentration in the spotted solution by dividing the droplet volume. For example, the linear range of 0.10 µL can be expressed as (6.9-160.0) × 10-8 M, and the LOD is 6.9 × 10-9 M.
and it has been proved that σ and Imax values would reduce if the oven temperature was lower than 60 °C. However, σ is a constant when the temperature is higher than 70 °C. We set the temperature at 70 °C in this assay. Calibration Curves. Figure 2 has proved that the variation of fluorescence intensity across the SOR belt section obeys a Gaussian distribution when the droplet volume is at the 0.x µL level. Based on that, we can establish following relationship between Imax and the amount of berberine (please see Supporting Information):
Imax )
K2
m ) ξm
4πσK1(KV1/3 - δ)
(4)
where K1 is an integral constant related to δ and σ, K2 is a constant related to the emission properties of the fluorescent materials, and m is the amount of the fluorescent materials. As K is related to the assembly conditions of SOR (see eq 2), thus ξ is a constant related to the emission properties of the fluorescent materials, evaporative velocity of the solvent, the droplet volume, and the properties of solution. As eq 4 shows, Imax is proportional to the amount of the fluorescent materials in the solution if V and σ are constant. In fact, it is very easy to adjust these variants related to the constant ξ by controlling pH, PVA concentration in the spotted solution, and the evaporation velocity of the solvent. Therefore, eq 4 constitutes the quantitative basis of the SOR technique. Table 1 displays the analytical parameters for the determination of berberine by using different droplet volumes. From which we can see that the present SOR method can be used to determine berberine in the range of 6.9 -160.0 fmol (or (6.9-160.0) × 10-8 M when the droplet volume is 0.10 µL), and the limit of determination reaches 0.69 fmol (or 6.9 × 10-9 M for a 0.10-µL droplet) with 3-fold signal-to-noise ratio (S/N ) 3). In addition, the slopes of the regression equations decrease with increasing droplet volume (under the same objective), proving the theoretical results of eq 4. Selectivity and Sample Determinations. Since the imaging of SOR displays the fluorescence properties of the drying specimen on the glass slide, which has a greatly minimized matrix effect compared to aqueous analysis, this assay demonstrates a high tolerance of foreign materials. It was found that common metal ions in urine could be allowed with a concentration level Analytical Chemistry, Vol. 74, No. 21, November 1, 2002
5567
Table 2. Determination Results of Berberine in Tabletsa
samples 1 2
content of berberine (mg/piece)
average (mg/piece, n ) 5)
RSD (%, n ) 5)
recovery (%, n ) 5)
28.25, 28.46, 28.07, 27.93, 28.50 28.81, 28.73, 28.51, 28.43, 28.31
28.24
0.85
98.88
28.56
0.74
99.14
a Reference value of tablet samples was 30 mg of berberine/piece supplied by the suppliers. Sample 1 was purchased from Kangfulai Pharmaceutical Ltd. (Chengdu, China), and sample 2 from Taiji Pharmaceutical Ltd. (Chongqing, China). Spotted solution: PVA-124, 0.06%; pH, 3.95. A 4× objective was used. Droplet volume, 1.00 µL.
Table 3. Determination of Recovery in Practical Urine Samplesa berberine (×10-4 mg/mL)b sample
found
added
total found
recovery (%)b
RSD (%)b
1 2 3
1.37 1.00 0.74
1.40 1.00 1.00
2.76 ( 0.04 1.99 ( 0.03 1.73 ( 0.02
97.14-105.00 97.00-104.00 97.00-101.00
3.2 2.8 1.5
a Urine samples were collected at 2 h after oral administration, and a 25-fold dilution was made before general analysis procedures. b Five measurements were made (n ) 5). Spotted solution: PVA-124, 0.06%; pH, 3.95. A 4× objective was used. Droplet volume, 1.00 µL.
above 1.0 × 10-5 M, ∼10-fold higher than the allowed levels reported by other authors 23(For details, see Supporting Information). Therefore, this method has good selectivity and can be applied to the direct determination of trace amounts of berberine in biological materials without prior separation of interfering species. Table 2 shows that the quantification results for berberine in tablets were in good agreement with reference values, indicating that the SOR method is reliable and practical. The purpose that we herein test the content of berberine in tablets, although its (35) Chen, M. Z. National Essential Drugs of ChinasWestern Medicines; People’s Health Press: Beijing, 1999; p 86. (36) Fang, W. X.; Song, C. S.; Zhou, L. X. Medicine Using Chinese Traditional Medicine Pharmacology, People’s Health Press: Beijing, 1998; pp 170-180.
5568 Analytical Chemistry, Vol. 74, No. 21, November 1, 2002
amount was at milligram level, is to identify the applicability of the present technique, and thus it can be applied to other determinations of berberine. Table 3 displays the determination results of berberine in human urine samples, and its recoveries of 97.00-105.00% and RSD of 1.5-3.2%. To further test the present method, we successfully monitored the excretion of berberine through human urine after oral administration. It was found that the average excretion rate of berberine can approach maximum within 1.5-2.5 h and almost reduce to zero during 6-7 h (see Supporting Information). This monitoring result is totally the same as other reports.35,36 CONCLUSIONS From above descriptions, we can see that the SOR technique has high sensitivity since the solutes concentrate on the ring of the droplet spot. Only nanoliter to microliter sample volumes are required; thus, this method can be used when sample size is minimal. Since this technique is based on solid deposition, the matrix effect in aqueous solution is avoided, allowing high tolerance level of coexisting substances. In addition, compared to other solid supports such as thin-film substrate, including octadecylsilanized silica and poly(vinyl alcohol) plate,3,4 the contribution from the background in this SOR technique is very small and can be made use of for sensitivity improvement. We believe that the SOR technique, if combined with a well-integrated mapping system, statistical analysis system, or a robot as the spotting engine to fully automate image collection, processing, and map construction, will become sufficiently general for various biochemical analysis and other analysis. ACKNOWLEDGMENT We thank the National Natural Science Foundation of China (NSFC, 20175017) and the University Key Teachers Program under the Ministry of Education of PRC (NO: [2000] 65-2804) for support of this research. SUPPORTING INFORMATION AVAILABLE Additional information as noted in the text. This material is available free of charge via the Internet at http://pubs.acs.org. Received for review July 11, 2002. Accepted September 6, 2002. AC0259358
