Double Input Capacitively Coupled Contactless Conductivity Detector

Article pubs.acs.org/ac

Double Input Capacitively Coupled Contactless Conductivity Detector with Phase Shift Hao Zheng,† Meng Li,† Jianyuan Dai,† Zhen Wang,† Xiuting Li,† Hongyan Yuan,*,‡ and Dan Xiao*,†,‡ †

College of Chemistry, Sichuan University, Chengdu 610064, People’s Republic of China College of Chemical Engineering, Sichuan University, Chengdu 610065, People’s Republic of China

‡

S Supporting Information *

ABSTRACT: A double input capacitively coupled contactless conductivity detector (DIC4D) device which gets higher sensitivity has been described in this paper. The detector consists of two input electrodes and one output electrode. When two alternating current (AC) voltages with the same amplitude and different phases are imposed on each input electrode, the equivalent resistance of the output electrode is reduced because of the interference of the two signals with different phase angles. For a capacitively coupled contactless conductivity detector (C4D), the ratio of the response of KCl solution to that of distilled water is 1.6. However, for DIC4D, the ratio is 1.55 at a phase difference of 0° and increases to 1.8 at the phase difference of 170°, respectively. For C4D, the response of KCl solution is a linear function of the logarithm of concentrations from 10−5 M to 10−2 M, and the slope is 5.58. However, the slope of the response increases to 7.13 in DIC4D, and the limit of detection (LOD) of DIC4D is estimated to be 5 × 10−8 M. The slope of the three-way DIC4D is increased to 69.78. A flow injection device is employed for the evaluation of the applicability of DIC4D with the same range, and good reproducibility is confirmed through flow injection of the same solution 10 times. The relative standard deviation (RSD) is 0.7%, which demonstrates a promising application to capillary electrophoresis (CE).

T

passing through the solution. Thus, the effect of phase angle was studied in detail in this paper. Meanwhile, the length of the electrode and the gap, wave forms, and frequency were also optimized. Different structures of DIC4D employed in flow injection were studied as well. The results showed an enhanced peak to baseline ratio, a higher sensitivity, a good linear function, and reproducibility.

he conductivity detector (CD) is used as a common tool for quantitative determination, for different materials have different conductivities. Such detection can be in contact mode1−6 or contactless mode. The contactless detector is more applicable than the contact one because the electrodes in the contactless detector are separated from the solution and could be prevented from contamination. Therefore, the capacitively coupled contactless conductivity detector (C4D) has received considerable attention and is widely applied to capillary electrophoresis (CE),7,8 Microchip,9−14 Micro-HPLC,15−18 and dual detector design.7,19,20 Nowadays, many efforts have been made to improve the performance of the contactless conductivity detector. Conventional C4D comprises an input electrode and an output electrode. To enhance the sensitivity of C4D, the endto-end differential model,12,21 a thinner insulating layer of 30 nm,22 and a high actuator voltage of 500 V were reported.23 A new method, which is to add an amplifier circuit before the data acquisition system and personal computer,24−26 was recently applied to C4D. However, it cannot improve the sensitivity of C4D itself, the improvement of which needs a change of the structure of C4D. In this study, the structure of C4D was modified to improve the sensitivity of C4D. Both two-input electrodes and oneoutput electrode were applied in C4D, which was called double input capacitively coupled contactless conductivity detector (DIC4D). In DIC4D, it was found that the signals interfered with each other with the principle of wave interference after © 2014 American Chemical Society

■

EXPERIMENTAL SECTION Chemicals. Potassium chloride was purchased from Tianjin Kermel Chemical Reagent Co., Ltd. (China). Sodium chloride was obtained from Tianjin Kermel Chemical Reagent Co., Ltd. (China). All reagents were of analytical grades unless otherwise stated. Ultrapure water was employed throughout. All stock solutions were kept in a refrigerator at 25 °C. All reagents unless otherwise stated were purchased from Aladdin Chemistry Co. Ltd. (Shanghai, China). Apparatus. The structure of the experimental device and the arrangement of the electrodes are illustrated in Figure 1. Output electrode 4 was placed in the middle, and the input electrodes 3 and 3′ were placed at both ends (Figure 1). The length, width, shape, and material of the three electrodes were all the same. The gaps L1 and L2 were of the same length. The glass tube with an outer diameter of 5 mm, inner diameter of Received: March 20, 2014 Accepted: September 24, 2014 Published: September 24, 2014 10065

dx.doi.org/10.1021/ac501199e | Anal. Chem. 2014, 86, 10065−10070

Analytical Chemistry

■

Article

RESULTS AND DISCUSSION C4D was widely employed alone or in combinations.27−34 In C4D, the degradation of input signal is different for different solutions. Since the solutions work as dielectric, and different solutions have different relative permittivities, this results in different output signal amplitudes. The input signal would weaken or strengthen when another signal was imposed at the same time according to the principle of wave interference, which was caused by phase change. In this work, a spectrum analyzer was employed to analyze the response difference between C4D and DIC4D. A 5 mM KCl solution was used as a sample solution. The frequency was 200 kHz. The length of the electrode was 5 mm. The length of the gap was 2 mm, and distilled water was used as the background electrolyte. In DIC4D, the response of the KCl solution with phase change was compared with that of distilled water. Optimizations were discussed in detail. In C4D, the response of 5 mM KCl was 4 dBm higher than that of distilled water, and the ratio of the response was 1.60 by calculating with eq 1.

Figure 1. Block diagram of the double input capacitively coupled contactless conductivity detector. (1) Function generator with two channels. (2) Glass tube used for detection. (3) Input electrode 3 and 3′. (4) Output electrode. (5) Data acquisition system. (6) Channel 1 of the function generator. (7) Channel 2 of the function generator. L1 and L2 are the length of the gap.

3.5 mm and total length of 7 cm was used (Wilmad, USA). The input and output electrodes were prepared by tightly winding copper foil (the thickness is 0.15 mm) around the glass tube. The input signals of 6 V (peak to peak) were actuated by a function generator (DG1022 RIGOL Technologies, Inc.). It had two channels of generator which were imposed on both input electrodes. One channel can generate a signal of 20 VPP while the other generates 6 VPP; thus 6 VPP was chosen in this work. A spectrum analyzer (DSA815 RIGOL Technologies, Inc.) was used to analyze the signals. The output signal was collected by a high frequency millivoltmeter transformer (HFJ8A, Shanghai Wuyi Electronics Co., Ltd.) and then recorded by a data acquisition system N2000 (Zhejiang University Zhida Information Engineering Co., Ltd.). The three-way DIC4D is illustrated in Figure S-1. Its outer diameter is 3 mm; the inner diameter is 1.5 mm. The three angles of the detector are 90°, 90°, and 180°; 45°, 135°, and 135°; and 120°, 120°, and 120°, respectively. The detector had also been studied in two approaches. One approach is that one branch of the detector is the input port, and two other branches are the output ports. The other approach is that one branch is blocked and filled with distilled water or air, one branch is the input port, while the other branch is the output port.

dBm = 20 × lg(V /V ′)

(1)

dBm is the unit of the signal strength. V is the response of the KCl solution, and V′ is the response of distilled water. In DIC4D, first, KCl solution and distilled water were measured at a phase difference of 0° (Figures S-2A, S-2B). The difference between the response of 5 mM KCl and distilled water is similar to that of C4D with a ratio of the response of KCl solution to water of about 1.55. Second, they were measured at a phase difference of 170° (Figures S-2C, S-2D). The result showed that the response of 5 mM KCl was nearly 5.2 dBm higher than that of distilled water. The ratio of the response of KCl solution to water increased to 1.80 in DIC4D. Obviously, the sensitivity of DIC4D was improved. In addition, the data acquisition system was used to analyze the response and the ratio of the response of DIC4D. The change of phase angle can be performed on one electrode while keeping the other one invariant, because the two input electrodes were symmetric. The responses of 5 mM KCl and distilled water were 264.6 and 171.0 mV at 0°, respectively, which are much higher than those of 170° (21.0 mV and 11.7 mV; Figure 2A). The decrease of the response indicated that the wave interference principle played an important role in the signal transmission process. In Figure 2B, the x-axis represented

Figure 2. Response and the ratio of the response of DIC4D at different phase angles. (A) Distilled water and 5 mM KCl have the same trend; at 0°, the response is the highest, and at 180°, the response is the lowest. (B) The ratio is the highest at 170° and the lowest at 0° in DIC4D. The ratio of C4D is nearly 1.6 and is kept the same as the phase angle changes. 10066



Article

the phase angle of one electrode from −170° to +170°, while the y-axis represented the value of the ratio. The ratio of the response at 170° was the highest and at 0° was the lowest. The equivalent circuit of DIC4D with a phase angle of 120° was the same as that of C4D. More experimentation was carried out, and it demonstrates that the structure of DIC4D and the phase shift together play an important role in improving the sensitivity. The effect of the electrode position on the response was also investigated. The input electrodes 3 and 3′ were placed at one end while the output electrode 4 was placed at another end (Figure S-3). The response of this setup was also measured with all the same aspects mentioned above. However, the ratio of the response was only 1.71. In this asymmetric device, the signals first interfered with each other at input electrode 3′ when they transmitted through the solution. It was different from the symmetric one, where signals interfered at the output electrode 4. A different electrode position results in a different length of the solution. In the symmetric device, lengths of the solution were the same. However, in the asymmetric one, lengths of the solution were L1 + L2 + l and L2, respectively, which led to some uncertain factors, such as it being difficult for the signals to interfere with each other due to the difference of the solution length. In this case, many other aspects of DIC4D need to be investigated, such as a frequency above 100 MHz, electrode materials, and so on. For simplicity, the symmetric device was chosen in the following study. An equivalent circuit of DIC4D is the combination of seriesparallel C4D detectors at resonant frequency (Figure 3), and

where R, X, and |Z| are the resistance (real part), reactance (imaginary part), and magnitude of the impedance of the C4D, respectively. The effects of the length of the electrode, the gap, and different solutions on AC impedance characteristics have been figured out in previous works.38 The impedance of C1 (ZC1) can be expressed as ZC1 = −

j j =− 2πfC1 2πf (2πεg εal)/ln(R /r )

(3)

where C1 is the capacitance of the electrodes, j = (−1)1/2, f is the frequency, l is the length of the electrodes, εa is the relative permittivity of vacuum, εg is the relative permittivity of the glass tube, R is the outer radius of the tube, and r is the internal radius of the tube. The impedance of C2 (ZC2) can be expressed as ZC2 = −

j j j =− =− 2πfC2 2πfεc 2εa(S /L) 2πfεc 2εa((πr 2)/L) (4)

where C2 is the capacitance of the solution, εC2 is the relative permittivity of the solution, εa is the relative permittivity of vacuum, r is the internal radius of the glass tube, S is the area of the cross section of the solution, and L is the length of the gap between electrodes. ZC2 is much larger compared with R2 when the frequency is relatively low. The impedance of the C4D (ZC4D) can be expressed as Z C4 D =

R2 + RD 1 + (R 2ωC2)2 ⎡ ⎢ 2 R 2 2ωC2 − j⎢ + C ⎢⎣ ωC1 1 + R 2 2ω 2 21 + C2

(

⎤ ⎥ 2⎥ ⎥⎦

)

(5)

where ω = 2πf, R2 is the solution resistance, and RD is the resistance of the detector circuit. The peak to baseline ratio of C4D (RAC4D) can be described as RA C4D =

IS IB R 2B 2

1 + (R 2BωC 2B)

=

Figure 3. Equivalent circuit of the DIC4D system. (A) Equivalent circuit of DIC4D. C1 is the input capacitance of both inputs and the output. R2 is the resistance of the solution between the electrodes. C2 is the capacitance of the solution. R3 is the resistance of the measuring part. (B) Signal interfere rule of DIC4D. Three phase angles, 0°, 120°, and 180°, are set as examples.

R 2S 1 + (R 2SωC 2S)2

V = |Z |

R + X2

2

where IS and IB are the current of the sample solution and background electrolyte, respectively, and R2S and R2B are the resistance of the sample solution and background electrolyte, respectively.35,39 On the basis of the equivalent circuit model of DIC4D in Figure 3A, the amplitude of the response is twice that of one signal when the two input signals are of the same phase. The result is almost the same as C4D when the two input signals have a phase difference of 120°. And the result is zero when they have a phase difference of 180°. It is obvious that the equivalent impedance of both the output electrode and the measuring circuit of DIC4D are reduced. The proportion of the

V 2

⎤ ⎥ C 1 + R 2B 2ω2( 21 + C 2B) ⎥ ⎦ ⎤ R 2S2ωC 2S ⎥ 2 C 1 + R 2S2ω2( 21 + C 2S) ⎥ ⎦ R 2B 2ωC 2B

(6)

lots of work has been done to analyze the equivalent circuit.35−37 In order to further demonstrate the advantages of the device that the sensitivity of DIC4D was improved, related formulas were derived. Physically, signal current (I) is detected when an AC voltage (V) is applied to the input detector.38 I=

⎡ 2 + R D − j ⎢ ωC + ⎢⎣ 1 ⎡ 2 + R D − j ⎢ ωC + ⎢⎣ 1

(2) 10067



Article

solution impedance is increased. The current of DIC4D (ID) can be expressed as

solution is larger in proportion in ZDIC4D than that in ZC4D. What’s more, when a capillary is employed, r is smaller than a glass tube. The resistance of R2 is much larger, which results in a higher ratio. In this work, four aspects were investigated to get higher sensitivity of the DIC4D. The first aspect was frequency. Response of the detector strongly depended on the excitation frequency. The impedance of the electrode capacitor and solution capacitor were reduced when the working frequency increased from 50 kHz to 10 MHz. The phase difference between the two input signals was set to 170° in detection. Figure S-4 showed that the ratio was the highest when the frequency was 200 kHz. When the frequency was higher than 5 MHz, the signal measured was higher but the ratio became smaller. Thus, a frequency higher than 10 MHz is not investigated in this work. The performance of DIC4D at higher frequencies is limited by two factors. One is that the bypass current increases while the operating frequency increases. Another is that the ratio declines, for the resistance of the solution accounts for a smaller proportion at higher frequency. Overall, the working frequency of 200 kHz was optimal and employed in the detection.37,40,41 The second aspect was the lengths of electrodes. Capacitance of the electrode is affected by the length of the electrode. When a longer electrode is employed, the proportion of the solution impedance is smaller, which leads to a lower sensitivity. The response ratio of different lengths of electrodes is shown in Table 1. Lengths of 2, 5, 8, and 10 mm for the electrodes were

⎛ 2π ⎞ ⎛ 2π ⎞ → ⎯ ID = I1⃗ + I2 = a sin⎜ t + φ1⎟ + a sin⎜ t + φ2⎟ ⎝T ⎠ ⎝T ⎠ Δ

φ2 − φ1→ δk

===== ⇒ k*a sin(0.5(φ2 − φ1)) = δka

(7)

where I1 and I2 are the currents of both input electrodes, respectively, a is the value of the current, φ1 and φ2 are the phases of the current of two input electrodes, k is the coefficient of the output current value, and δk is the coefficient of the current. Note that I1 and I2 in the formula are vectors. A dielectric phase shift can be neglected because the shifter works at frequencies around 5 GHz, the frequency which belongs to the microwave. The equivalent impedance of DIC4D (ZDIC4D) can be expressed as Z DIC4D =

R2 + δ kR D 1 + (R 2ωC2)2 ⎡ ⎢δ + 1 R 2 2ωC2 − j⎢ k + δC ⎢ ωδkC1 1 + R 2 2ω 2 δ k+11 + C2 ⎢⎣ k

(

⎤ ⎥ ⎥ 2 ⎥ ⎥⎦

)

(8) 4

The peak to baseline ratio of DIC D (RADIC D) can be described as IS IB → ⎯ I + = ⎯→s1 Ib1 +

4

Table 1. Optimization of Detection Parameters of DIC4Da

RADIC4D =

R 2B 1 + (R 2BωC 2B)2

= R 2S 1 + (R 2SωC 2S)2

frequency: 200 kHz

⎯→ Is2 ⎯→ Ib2

⎡ δ +1 + δkRD − j⎢ ωδk C + ⎢ k 1 ⎣ ⎡ δ +1 + δkRD − j⎢ ωδk C + ⎢ k 1 ⎣

parameter

⎤ ⎥ 2 δ C 1 + R 2B 2ω2( δ k+11 + C 2B) ⎥ ⎦ k ⎤ R 2S2ωC 2S ⎥ 2 δ C 1 + R 2S2ω2( δ k+11 + C 2S) ⎥ ⎦ k R 2B 2ωC 2B

length of gap (mm)

wave form

(9)

where IS1 and IS2 are the currents of the sample solution and IB1 and IB2 are the currents of the background electrolyte. R2S and C2S are the resistance and the capacitance of the sample solution; R2B and C2B are the resistance and the capacitance of the background electrolyte. Then, the ratio of RADIC4D and RAC4D can be obtained as follows:

employed. When the length was larger, the proportion of the solution impedance increased and the ratio was higher. Recent researches showed that lengths of 2 mm and 5 mm resulted in better performance. For the convenience of experiments, the conventional length of the electrode (5 mm) was chosen. The third aspect was the length of the gap between electrodes. A different length of the gap results in a different length of the solution. The impedance of the solution depends not only on the electrical conductivity and the cross-sectional area but also on the length of the solution. Impedance of the solution was larger when the length of the gap was longer. The ratio of the response of different lengths of the gap between electrodes is shown in Table 1. When the gap was shorter, the solution impedance was smaller, and the proportion of the solution impedance and the impedance of the whole device

(ZC2 //Z R 2)S + ZC1 + δk(ZC1 + Z D) (ZC2 //Z R 2)B + 2ZC1 + Z D (ZC2 //Z R 2)S + 2ZC1 + Z D

1.42 1.81 2.15 2.22 1.81 1.81 1.80 1.80 1.81 1.72 1.51 1.48

Optimization of length of electrode, length of gap, and wave forms. Frequency: 200 kHz.

(ZC2 //Z R 2)B + ZC1 + δk(ZC1 + Z D)

/

2 5 8 10 2 5 8 10 sine triangle square pulse

a

RADIC4D/RA C4D ≈

ratio of KCl soltion to water

length of electrode (mm)

(10)

The effective range of δk is 0 < δk < 2. The value of δk depends on the phase difference φ2 − φ1. Equation 10 shows that when the phase angle is between 0° and 120°, δk is larger than 1, and RADIC4D is smaller than RAC4D. When the phase angle is between 120° and 180°, δk is smaller than 1, and RADIC4D is larger than RAC4D. Meanwhile, the impedance of the 10068



Article

were also smaller, leading to a smaller ratio and a higher measured voltage. The ratio was the smallest when the gap was 2 mm, which accordingly was employed in further experiments. In addition, four different wave forms were studied in this work. As shown in Table 1, the ratio of the response of the sine wave was the highest, while the measured voltage was the lowest. The triangle wave exhibited a slightly lower ratio and higher voltage. However, the square wave and the pulse presented a similar low ratio and high voltage. In Figure S-5, the signal strength of the output sine wave was the weakest. The strength of the output triangle wave was stronger than that of the sine wave. The strength of the square wave was the strongest. The pulse is kept constant after signals interfere with each other, and the ratio of the pulse is the lowest. Therefore, the sine wave was employed in the study. The KCl solution in a concentration range from 5 × 10−8 M to 0.1 M was employed to investigate the response of DIC4D. As shown in Figure 4, the response range was larger than that

Figure 5. Responses of DIC4D and C4D in a glass tube and capillary. The response of DIC4D at 170° and the response of C4D of the concentration from 10−4 M to 10−1 M were compared after normalization.

solution went among all three electrodes, the solution diffused in the process, and the peak was lower than that of C4D. Then, it was carried out with a capillary with an inner diameter of 1 mm, and the response is shown in Figure 5B. Compared with the glass tube, it took much less time to run from electrode 3 to 4 because of the small inner diameter of the capillary. In this short time, the solution could not diffuse during the whole process in which the solution went among all three electrodes, which resulted in normal peaks. The response of the three-way DIC4Ds had been studied, and it demonstrates that the detector with angles 120°, 120°, and 120° had better performance than others. Figure S-7 shows the response of C4D and a three-way DIC4D with angles 120°, 120°, and 120° after normalization. One branch of the DIC4D is blocked. It is obvious that the ratio is highly increased. To further evaluate the performance of the three-way DIC4D, the response of KCl solution with a concentration range was studied. In Figure S-8, the response of C4D and three-way DIC4D with a blocked branch filled with air were studied. The slope of the three-way DIC4D is 69.78, while that of C4D is 13.70. The sensitivity of the three-way DIC4D was significantly improved. When one branch is blocked and filled with air, another branch is filled with the KCl solution; the input signals were completely different. As distilled water is running, the output signal is very small because the electrodes are symmetric. As KCl solution is running, the output signal is received by the signal from the electrode filled with KCl solution minus that from the electrode filled with air. The baseline is largely decreased while the peak height is nearly the same, thus highly increasing the ratio. The blocked branch with water was also studied; the performance is worse than that filled with air, for the water would be contaminated while the KCl solution is running. The responses of 15 different solutions were studied (Figure S-9). The concentration of them was 0.01 M. It is obvious that the response of KH2PO4 solution was the lowest, and the responses of BaCl2 and Fe(NO3)3 solutions were the highest. It is because the electrical conductivity of KH2PO4 solution was the lowest and that of BaCl2 and Fe(NO3)3 solutions were the highest. In addition, 5 mM KCl was measured 10 times in a capillary with DIC4D. The relative standard deviation of the responses of KCl solution was 0.7% (Figure S-10). A good reproducibility of DIC4D is obtained. Advantages of simple

Figure 4. Response curves of C4D and DIC4D. Response of the concentration of KCl solution ranging from 5 × 10−8 M to 0.1 M in DIC4D. The linear range is between 10−5 M and 10−2 M, and its correlation coefficient is 0.99.

of C4D. LOD of DIC4D is estimated to be 5 × 10−8 M. The slope of the response of DIC4D is also higher than that of C4D. Obviously, the sensitivity of DIC4D was significantly improved. The response was a linear function of the logarithm of the concentration at concentrations from 10−5 M to 10−2 M. It is obvious that the slope of DIC4D (7.13) is higher than that of C 4 D (5.58) after normalization. Both the correlation coefficients were 0.99. Different input voltages from 3 VPP to 20 VPP were employed. In Figure S-6A, it shows the ratio of response of DIC4D is always higher than that of C4D as the input increases. In Figure S-6B, it is obvious that the sensitivity improves as the input voltage increases, and the slopes of DIC4D are always higher than those of C4D with the same input. To further evaluate the reliability of DIC4D, the flow injection device has been employed in this work. First, it was carried out with the glass tube. The response is shown in Figure 5A. The concentration of KCl solution ranged from 10−4 M to 10−1 M. It is obvious that there is a pulse at the front of the peak, which is because it took time for the solution to run from electrode 3 to electrode 4 (Figure 1) when it went through the glass tube. The responses of the two input signals were completely different during the process. Thus, the pulse was produced because the two signals could not interfere with each other perfectly, leading to a significantly enhanced signal peak, which played a key role in improving the sensitivity. When the 10069



Article

(11) Liu, B.; Zhang, Y.; Mayer, D.; Krause, H.-J.; Jin, Q.; Zhao, J.; Offenhäusser, A. Electrophoresis 2011, 32, 699−704. (12) Fercher, G.; Haller, A.; Smetana, W.; Vellekoop, M. J. Anal. Chem. 2010, 82, 3270−3275. (13) Zemann, A. J. Electrophoresis 2003, 24, 2125−2137. (14) Tomazelli Coltro, W. K.; Fracassi da Silva, J. A.; Carrilho, E. Anal. Methods 2011, 3, 168−172. (15) Pumera, M. Talanta 2007, 74, 358−364. (16) Kubáň, P.; Hauser, P. C. J. Chromatogr. A 2007, 1176, 185−191. (17) Haddad, P. R.; Nesterenko, P. N.; Buchberger, W. J. Chromatogr. A 2008, 1184, 456−473. (18) Wu, J.; Zhang, H.; Qiu, J. J. Hazard. Mater. 2012, 215−216, 138−145. (19) Novotný, M.; Opekar, F.; Jelínek, I.; Štulík, K. Anal. Chim. Acta 2004, 525, 17−21. (20) Tan, F.; Yang, B.; Guan, Y. Anal. Sci. 2005, 21, 583−585. (21) Huang, T.; Kang, Q.; Zhu, X.; Zhang, Z.; Shen, D. Anal. Methods 2013, 5, 6839−6847. (22) Guijt, R. M.; Baltussen, E.; van der Steen, G.; Frank, H.; Billiet, H.; Schalkhammer, T.; Laugere, F.; Vellekoop, M.; Berthold, A.; Sarro, L.; van Dedem, G. W. K. Electrophoresis 2001, 22, 2537−2541. (23) Tanyanyiwa, J.; Hauser, P. C. Anal. Chem. 2002, 74, 6378−6382. (24) Kubáň, P.; Karlberg, B.; Kubáň, P.; Kubáň, V. J. Chromatogr. A 2002, 964, 227−241. (25) Mayrhofer, K.; Zemann, A. J.; Schnell, E.; Bonn, G. K. Anal. Chem. 1999, 71, 3828−3833. (26) Tan, F.; Yang, B.; Guan, Y. Anal. Sci. 2005, 21, 955−958. (27) Zhang, L.; Khaloo, S. S.; Kubán, P.; Hauser, P. C. Meas. Sci. Technol. 2006, 17, 3317−3322. (28) Kappes, T.; Hauser, P. C. J. Chromatogr. A 1999, 834, 89−101. (29) Pacáková, V.; Coufal, P.; Štulík, K. J. Chromatogr. A 1999, 834, 257−275. (30) Polesello, S.; Valsecchi, S. M. J. Chromatogr. A 1999, 834, 103− 116. (31) Kubáň, P.; Hauser, P. C. Electrophoresis 2011, 32, 30−42. (32) Kubáň, P.; Hauser, P. C. Electrophoresis 2009, 30, 176−188. (33) Guijt, R. M.; Evenhuis, C. J.; Macka, M.; Haddad, P. R. Electrophoresis 2004, 25, 4032−4057. (34) Coltro, W. K. T.; Lima, R. S.; Segato, T. P.; Carrilho, E.; de Jesus, D. P.; do Lago, C. L.; da Silva, J. A. F. Anal. Methods 2012, 4, 25−33. (35) Shen, D.; Li, D.; Yang, X.; Zhu, Y.; Dong, J.; Kang, Q. Talanta 2011, 84, 42−48. (36) Zemann, A. J.; Schnell, E.; Volgger, D.; Bonn, G. K. Anal. Chem. 1998, 70, 563−567. (37) Kubáň, P.; Hauser, P. C. Electrophoresis 2004, 25, 3387−3397. (38) Chang Guo, C.; Lei Guang, L.; Yu Jun, S.; Yu Ping, L. Electrochim. Acta 2009, 54, 6959−6962. (39) Brito-Neto, J. G. A.; Fracassi da Silva, J. A.; Blanes, L.; do Lago, C. L. Electroanalysis 2005, 17, 1198−1206. (40) Opekar, F.; Štulík, K.; Fenclová, K. Electroanalysis 2010, 22, 161−167. (41) Gaš, B.; Zuska, J.; Coufal, P.; van de Goor, T. Electrophoresis 2002, 23, 3520−3527.

structure, low-cost, stability, and highly sensitive detection are shown in this device. It will provide a new strategy for miniaturized instruments in ion analysis, and the detector has the potential to be applicable to conventional capillary electrophoresis (CE) and ion chromatography (IC) system.

■

CONCLUSION This paper proposed a new device (DIC4D), the structure of which was modified from C4D to improve the sensitivity, and it clearly demonstrates the advantages of DIC4D. The three pieces of copper foil around the glass tube, which act as three cylindrical capacitors, allows a fast adjustment of the DIC4D on virtually any position. The impedance of the output electrode was equivalently reduced. The ratio of the response of C4D was 1.6, and that of DIC4D increased to 1.8 at 170°. Optimization of the working frequency showed that medium frequencies around 200 kHz had the highest ratio, and medium frequency caught the most attention in C4D for its good performance, which means that the presented DIC4D kept the advantage of C4D. The DIC4D was certified to have higher sensitivity, good stability, and reproducibility, which gives it the potential to be applied to capillary electrophoresis, ion chromatography, and other devices in the future. Furthermore, the new design may further improve the sensitivity of DIC4D and its potential for real application.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional material as described in the text. This material is available free of charge via the Internet at http://pubs.acs.org/.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Financial supports from National Natural Science foundation of China (No. 21377089 and No. 21177090) are gratefully acknowledged.

■

REFERENCES

(1) Avdalovic, N.; Pohl, C. A.; Rocklin, R. D.; Stillian, J. R. Anal. Chem. 1993, 65, 1470−1475. (2) Harrold, M.; Stillian, J.; Bao, L.; Rocklin, R.; Avdalovic, N. J. Chromatogr. A 1995, 717, 371−383. (3) Huang, X.; Luckey, J. A.; Gordon, M. J.; Zare, R. N. Anal. Chem. 1989, 61, 766−770. (4) Huang, X.; Pang, T. K. J.; Gordon, M. J.; Zare, R. N. Anal. Chem. 1987, 59, 2747−2749. (5) Huang, X.; Zare, R. N. Anal. Chem. 1991, 63, 2193−2196. (6) Huang, X.; Zare, R. N.; Sloss, S.; Ewing, A. G. Anal. Chem. 1991, 63, 189−192. (7) Chvojka, T.; Jelínek, I.; Opekar, F.; Štulík, K. Anal. Chim. Acta 2001, 433, 13−21. (8) Kaniansky, D.; Zelenská, V.; Masár, M.; Iványi, F.; Gazdíková, Š. J. Chromatogr. A 1999, 844, 349−359. (9) Abad-Villar, E. M.; Tanyanyiwa, J.; Fernández-Abedul, M. T.; Costa-García, A.; Hauser, P. C. Anal. Chem. 2004, 76, 1282−1288. (10) Ansari, K.; Ying, J. Y. S.; Hauser, P. C.; de Rooij, N. F.; Rodriguez, I. Electrophoresis 2013, 34, 1390−1399. 10070


Double Input Capacitively Coupled Contactless Conductivity Detector

Recommend Documents