Flow Through Horizontal Tubes Submerged in Water

Subscriber access provided by BALL STATE UNIV
Article
Flow through horizontal tubes submerged in water in the
absence of a pressure gradient: mechanistic considerations
Mina Rohani, and Gerald H. Pollack
Langmuir, Just Accepted Manuscript • DOI: 10.1021/la4001945 • Publication Date (Web): 30 Apr 2013
Downloaded from http://pubs.acs.org on May 5, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Page 1 of 24 Langmuir
1
2
3
4
5
6
7
8
Flow through horizontal tubes submerged in water in
9
10
11
12 the absence of a pressure gradient: mechanistic
13
14
15
16 considerations
17
18
19
20
21 Mina Rohani and Gerald H. Pollack
22
23
24
25
Department of Bioengineering, University of Washington, Seattle, WA 98195, United States
26
27
28
29
30
31 ABSTRACT
32
33
34
35 A self-driven flow was observed in Nafion and other hydrophilic tubes immersed in water. The
36
37 intratubular flow was generated when water came in contact with the tube’s hydrophilic surfaces.
38
39 Flow characteristics were studied in tubes of varying size, exposed to light of different intensities
40
41
42 and wavelengths. The results lead to the hypothesis that the flow is driven by a high
43
44 concentration of protons accumulating inside the tube, creating an axial proton gradient between
45
46
47
the inside and outside of the tube. We also demonstrate a faster flow under incident light,
48
49 particularly at UV wavelengths, implying that proton generation may be driven by light.
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
1
Langmuir Page 2 of 24
1
2
3
INTRODUCTION
4
5
6
7 Fluid transport occurs in the presence of a driving force. This force may arise from a pressure
8
9 gradient, an electrokinetic or magnetohydrodynamic drive, or various other phenomena
10
11
12 (Castellanos, 1998) (Davidson, 2001). In any case, flow results from the conversion of any of
13
14 various types of energy into kinetic energy.
15
16
17 Immersing Nafion in water with microspheres results in formation of a microsphere-free zone at
18
19
20 the interface of the hydrophilic material and water (Zheng et al, 2006). Since this zone excludes
21
22 various particles and solutes, it has become known as the “exclusion zone” or EZ. Many earlier
23
24
studies have shown that the water adjacent to hydrophilic surfaces has different characteristics
25
26
27 than bulk water, both physically and chemically (Szent-Gyorgyi, 1957) (Ling, 1984). These
28
29 hydrophilic surfaces include Nafion and various gels (Zheng et al., 2006).
30
31
32
33
When Nafion tubes are immersed in water, water flows continuously through the tube. Since
34
35 Nafion also bears exclusion zones, we wished to test whether the flow generation was related to
36
37 the EZ, and further, what might be the flow-driving mechanism.
38
39
40
41
42
43
44 EXPERIMENTAL METHODS
45
46
47
48
Microsphere Suspension
49
50
51 Nafion tubes (Perma Pure LLC) were submerged in distilled, deionized water (Nanopure,
52
53 Barnstead, 18.1 MΩ cm). In order to visualize the flow inside the tubes, Polybead carboxylate 1-
54
55
56 µm microspheres (Polyscience Inc.) were suspended in the water, at a microsphere-to-water
57
58
59
60
2
1
2
3
volume ratio of 1 to 300. The volume ratio was kept constant in all experiments to eliminate any
4
5
6 effects that might arise out of concentration differences.
7
8
9 Setup
10
11
12
13 A standard chamber was used for the experiments unless otherwise noted. The chamber was
14
15 made from a rectangular polycarbonate block with a hollow cylinder cut out (8.7 mm diameter,
16
17 5.8 mm high; volume 350 µl). The bottom was covered with a glass slide and the top was left
18
19
20 uncovered. The Nafion tube was positioned horizontally at the bottom of the chamber and held
21
22 by a thin barrier.
23
24
25
Figure 1 shows the chamber. Panel a shows an exploded schematic, while panel b shows the
26
27
28 assembled chamber, ready to be filled with water. The thin vertical barrier, fixed with screws,
29
30 has a small cut underneath for holding the tube at the bottom of the chamber. The barrier does
31
32
33
not separate the chamber into two parts, as water can flow freely above and below the barrier.
34
35
36
37
38
39
40
41 Assembled:
42
43
44
45 (a) (b)
46
47 Figure 1. Experimental chamber. The chamber is 5 cm long, 2.5 cm wide and 5.5 mm high. A
48
49 glass slide is glued to the bottom. The cylindrical cavity has a diameter of 8.8 mm, a height of
50
51 5.8 mm, and a volume of 350 µl.
52
53
54
55 Nafion becomes hydrophilic as it is hydrated (Goswami, Klaus, & Benziger, 2008). Prior to all
56
57 experiments, the Nafion was allowed to prehydrate for 15 minutes. Each test started with a new
58
59
60
3
1
2
3
pre-washed Nafion tube. All water was taken from a Barnstead Diamond ultrapure source. The
4
5
6 water showed some pH variation for approximately half an hour after dispensing. In a
7
8 representative test, the pH of the water, measured with a Fisher Scientific pH meter, was 5.4
9
10
11
immediately after leaving the Barnstead unit, 6.0 after 10 minutes, 6.4 after 20 minutes and then
12
13 formed a plateau at pH of 6.6 by 30 minutes and afterwards. The pH increase was found
14
15 consistently, and may be tied to carbon dioxide exchange. In order to avoid any effects of such
16
17
18 pH variation, the water was left to stand in the chamber for approximately 30 minutes prior to
19
20 experimentation.
21
22
23 Figure 2 shows a cross-sectional picture of a hydrated Nafion tube immersed in a microsphere
24
25
26 suspension. The image shows annular, microsphere-free EZs both inside and outside the tube.
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47 Figure 2. Cross sectional view of a Nafion tube in an aqueous microsphere suspension. The
48
49
50 brighter white areas show the exclusion zones, from which microspheres have been expelled.
51
52
53 An inverted Nikon Diaphot microscope with an attached video camera (Edmund Optics)
54
55 captured the flow inside the tube. The microscope’s light intensity was kept constant in all
56
57
58 experiments in order to guard against any inadvertent light effects on flow.
59
60
4
1
2
3
Flow characteristics were studied for tube lengths ranging from 1 mm to 5 mm, and two tube
4
5
6 diameters. The tubes’ outer diameters (dry) were 840 µm and 1,500 µm. When hydrated, Nafion
7
8 tubes absorb water and expand. The final outer diameters were approximately 925 µm and 1,800
9
10
11
µm respectively. Hydrated inner diameters were 725 µm and 1,400 µm. Tube characteristics,
12
13 including both diameter and length are summarized in Table 1.
14
15
16 Velocity values were measured by recording videos of microspheres flowing through the tube,
17
18
19 and using ImageJ software for particle tracking. Maximum velocities generally occurred along
20
21 the tubes’ central axis. Throughout this paper, the values reported were those central, maximum
22
23 values. An automatic program recorded the velocity every six minutes, starting at two minutes
24
25
26 after immersion in the microsphere suspension, two minutes being the average time required for
27
28 full EZ formation. EZ sizes were measured from captured images, using a reference scale.
29
30
31
32 Tube
Tube D [mm] L [mm] D [mm] L [mm]
33
34 Name Name
35 D1L1 0.84 1 D2L1 1.5 1
36
37 D1L2 0.84 2 D2L2 1.5 2
38 D1L3 0.84 3 D2L3 1.5 3
39
40 D1L4 0.84 4 D2L4 1.5 4
41 D2L5 1.5 5
42
43
44 Table 1. Tubes categorized according to diameters (D1 and D2) and lengths (L1 to L5).
45
46
47 Light Source
48
49
50
51
The microscope’s halogen lamp was used as the white light source. Light intensity was
52
53 adjustable, and measured using a power meter (Newport- model 1815C). The chamber was
54
55 illuminated as uniformly as possible in order to avoid local heating. For basic experiments the
56
57
58 power of the white light was kept constant at 1 mW.
59
60
5
1
2
3
For the experiments using a UV source, a custom-made broadband UV-Vis lamp (wavelength
4
5
6 range 332 - 392 nm) was used. In order to study the transient impact of wavelength change on
7
8 flow, a protocol was adopted to alternate between white and UV light sources, maintained at
9
10
11
equal intensity. The sources were alternated every 20 minutes, approximately.
12
13
14 Gel Preparation
15
16
17 In addition to Nafion tubes, we also tested polyethylene glycol (PEG) channels and polyvinyl
18
19
20 alcohol (PVA) tubes. To make PEG microchannels, a premade SU-8 mold was used. The mold
21
22 was a simple single rectangular microchannel mold (2 mm width, 10 µm height, and 20 mm
23
24
length). PEG was prepared from a low molecular weight (250) Poly(ethylene glycol) diacrylate
25
26
27 (Sigma Aldrich), mixed with 2-hydroxy-2-methylpropiophenone (Sigma Aldrich) to a volume
28
29 ratio of 2% (98% PEG), and poured onto the mold in a petri dish. It was then polymerized under
30
31
32
UV light for 5 minutes. The polymerized PEG was transparent. The channel was left uncovered.
33
34
35 PVA was prepared by mixing polyvinyl alcohol with water at a ratio of 4%. For polymerization
36
37 purposes 4% borax was added. The volume ratio of the borax to PVA solution was 1 to 4. The
38
39
40 solution was then heated for 20 minutes at 80 ºC. Preparation required stirring until gelling was
41
42 complete. As the gel was not quite a moldable material, a conical tunnel was created, by using a
43
44 plastic cone as a mold (6 mm larger diameter, 2 mm smaller diameter, and 20 mm length). We
45
46
47 surrounded the cone mold with the gel and applied some pressure to form the PVA gel. Then, the
48
49 mold was removed and the tunnel-containing gel was immersed in water for testing.
50
51
52
53
54
55
56 RESULTS
57
58
59
60
6
1
2
3
We consistently observed flow through the Nafion tubes as well as polyethylene glycol (PEG)
4
5
6 channels and polyvinyl alcohol (PVA) tunnels. The observations reported below refer mainly to
7
8 the Nafion tubes. Most of the observations were made using the standard chamber (see
9
10
11
EXPERIMENTAL METHODS). Later, we found that flow could persist for much longer in
12
13 larger chambers. Those latter observations are described later in the results.
14
15
16 While a fixed microsphere concentration (volume ratio 1 to 300) was routinely used, lower
17
18
19 concentrations were tested, down to 10% of the standard concentration. We could find no
20
21 obvious differences in flow velocities (except that fewer microspheres made detection more
22
23 challenging), confirming that the presence of microsphere played no more than an incidental role
24
25
26 in quantification of results.
27
28
29 EZ Size
30
31
32
33
Before studying flow characteristics, we measured EZ sizes inside and outside the tube, at
34
35 random locations along the length. This was done before the flow started, during the course of
36
37 flow, and after the flow had stopped. Some variation in inside and outside EZ size was observed,
38
39
40 but we could detect no systematic changes. In one representative example, the inside EZ was
41
42 measured every 10 min, over 2 hours. Averaging over time gave a mean EZ size of 156 +/- 20
43
44 µm. The modest SD indicated that the EZ appeared to be reasonably stable over time.
45
46
47
48 “Success” Rate
49
50
51 Short tubes always resulted in flow regardless of tube diameter. For 1 and 2 mm tube lengths, of
52
53
54
both diameters, the success rate was 100% (n = 8).
55
56
57
58
59
60
7
1
2
3
D2L3 and D2L4 also showed a relatively high success rate of 86% (n = 7). Increasing the length,
4
5
6 D2L5 showed flow about 50% of the time (n = 6). D2 tubes with lengths longer than 5 mm
7
8 showed little or no flow.
9
10
11
12 Smaller diameter tubes generally showed lower success rates. For D1L3 the success rate was
13
14 50% (n = 6). D1L4 either generated a very slow flow (velocity was generally lower than 2 µm/s),
15
16 or no flow. Increasing the D1 length to 5 mm or longer resulted in no flow. In D1 tubes,
17
18
19 therefore, the chance of seeing flow was high only for lengths shorter than 4 mm.
20
21
22 Spontaneous Flow Behavior
23
24
25
Velocity Variations over Time
26
27
28
29 To study the flow pattern over time, 1.5 mm diameter, 4-mm long Nafion tubes (D2L4) were
30
31 used. The results from five tests are plotted in Figure 3. The flow generally started within the
32
33
34 first minute. Velocity values were highest in the first 5 to 10 minutes and then progressively
35
36 diminished over time.
37
38
39
40 Velocity Variation over Time
41 80
42 D2L4
43 70
44 D2: 1.5 mm – L4: 4 mm
Velocity [µm/s]
60
45
46 50
47
48 40
49 30
50
51 20
52 10
53
54 0
55 0 12 24 36 48 60 72 84 96 108 120
56 Time [min]
57
58
59
60
8
1
2
3
Figure 3. Flow velocity over two hours in 4-mm long Nafion tubes. Mean velocity (n = 5) in red,
4
5
6 +/- standard deviation in blue.
7
8
9 Velocity as a Function of Tube Diameter and Length
10
11
12
13 Generally, we found that fluid velocities increased as the tube length increased towards an
14
15 optimal value. For a D2 tube, the optimal length was 4 mm; for the D1 tube, it was 3 mm.
16
17
18 Regarding diameter variation at constant length, experiments showed much lower velocities for
19
20
21 the smaller diameter. For example, for an L4 length, maximum mean velocity was as large as 50
22
23 µm/s in the D2 tube, while in the D1 tube it did not exceed 7 µm/s.
24
25
26
27
Effect of Chamber Size on Flow Velocity and Duration
28
29
30 Preliminary observations had shown that larger chambers produced larger and longer-lasting
31
32 flows. In order to study these effects in detail, the total volume of fluid passing through the tube
33
34
35 until flow stoppage was measured in different size chambers. D2L3 tubes were used. The
36
37 smallest chamber, routinely used (Figure 1), had a volume of 350 µl, while the largest had a
38
39 volume of 6,000 µl. Knowing the tube diameter and flow velocity, we could calculate the total
40
41
42 volume of fluid that had passed through the tube during the entire time the flow was ongoing
43
44 (Figure 4). The numbers shown in the plot, therefore, represent the total liquid volume
45
46
transported until the flow stops. Because the stop times differ for the respective data points, the
47
48
49 flow durations differ as well.
50
51
52 We found that larger chambers resulted in larger fluid volumes. The highest volume occurred in
53
54
55 the 6,000-µl chamber, and that was also the longest lasting, typically 1.5 days. For smaller
56
57 chambers, including the standard one, the flow duration usually did not exceed 3 - 4 hours. The
58
59
60
9
1
2
3
effects of chamber size on the fluid volume were dramatic: increasing chamber volume from 350
4
5
6 µl to 6000 µm increased the volume transported by more than ten times, and the linearity of the
7
8 curve suggested that further increases of chamber size would continue to increase the volume
9
10
11
transported.
12
13
14
15 Total Volume of Fluid vs. Chamber Size
16 7000
17
18 6000
19
20 5000
Volume [µl]
21
22 4000
23
3000
24
25 2000
26
27 1000
28
29 0
30 0 1000 2000 3000 4000 5000 6000
31 Chamber Size [µl]
32
33
34
35
36 Figure 4. Volume of fluid passing through the tube (D2L3 tube) in various chamber sizes.
37
38 Averaged data (n = 3) in red, and +/- standard deviation in blue.
39
40
41
42 Flows in Gel Channels and Tunnels
43
44
45 We observed spontaneous flow in PEG microchannels and PVA tunnels. Further, both materials
46
47
showed exclusion zones qualitatively similar to those of Nafion (Figure 5). The PVA EZ was
48
49
50 largest, Nafion next, and PEG smallest. A tapered mold was used for PVA because it was easy to
51
52 remove; at the same time, the taper provided a means to check the flow direction: all the sample
53
54
55 cases (n = 3) showed flow from the larger to smaller diameter.
56
57
58
59
60
10
1
2
3
PVA also resulted in the highest velocity among all materials (sometimes up to 100 µm/s),
4
5
6 although the flow did not persist as long as in the Nafion tubes. PEG also led to stable flows
7
8 lasting up to several hours, although the velocities (typically 5 µm/s), were lower than Nafion
9
10
11
(typically 10 µm/s).
12
13
14
Microchannel wall
Microchannel wall
15
16 Microspheres Microspheres
17 EZ EZ EZ EZ
18
19
20
21
22 Flow Direction Flow Direction
23 Tunnel wall
24 (a) PEG Microchannel (b) PVA Tunnel
25
26
PVAmicro-channel (b).
Figure 5. Exclusion zone inside PEG micro-channels (a) and PVA conical
27
28 EZ boundary is shown by lines.
29
30
31
32 Flow Depends on Incident Light
33
34
35 Effects of Light Intensity
36
37
38
39 To study the effects of light on flow, we used the microscope’s lamp at a series of incident
40
41 intensities. The tube size (D2L3) and chamber size (350 µl) were fixed for this and subsequent
42
43 experiments on the effects of light. In each experiment, we measured the peak velocity (Figure
44
45
46 6). Results are shown for a representative run (a) and the average over multiple runs (b).
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11
1
2
3
4 Representative Peak Velocity vs. Light Power
5
50
6
Peak Velocity [µm/s]

7 40
8
9 30
10
11 20
12
13 10
14
15 0
16 0 500 1000 1500 2000
17 Light Power [µW]
18
19
20 (a)
21
22
23 Mean Peak Velocity vs. Light Power
24
25 50
Peak Velocity [µm/s]
26
27 40
28
29 30
30
31 20
32
33 10
34
35 0
36 0 500 1000 1500 2000
37
38 Light Power [µW]
39 (b)
40
41
42
43
Figure 6. Effect of white light power on peak velocity: (a) representative test; (b) mean peak
44
45 velocity for the respective run (n = 4). Means in red, and standard deviation in blue. A D2L3
46
47 tube was used in a 350-µl chamber.
48
49
50
51 Effects of Optical Wavelength
52
53
54 Previous studies have shown that EZ expansion depends on wavelength (Chai, Yoo, & Pollack,
55
56 2009). We therefore examined the effects of different wavelengths on flow. To do this, we
57
58
59
60
12
1
2
3
substituted a broadband UV source for the microscope’s white light source, with optical powers
4
5
6 set to be equal. Figure 7 shows fluid velocity over time in the respective situations. UV gave
7
8 higher velocities, but those velocities diminished more quickly than when the white light source
9
10
11
was used.
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33 Figure 7. Velocity variation over time with white light source and UV source, at same optical
34
35 power levels. Representative runs are shown. A D2L3 tubes was used in a 350-µl chamber.
36
37
38 In addition to the observations depicted in Figure 7, a special alternation protocol was used to
39
40
41 compare the effects of the two wavelength ranges. D2L3 tubes were used in the 350-µl chamber.
42
43 A protocol of 20-minute alternation was followed in five experiments with similar results. Figure
44
45
8 shows a representative example.
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
1
2
3
4 Velocity Variation with Alternating White and UV Sources
5
6 30
7
25
8
9 20
Velocity [µm/s]
10
11 15
12
13 10
14
15 5
16
17 0
18 0 50 100 150 200 250
19 -5
20 Time [min]
21
22
23
24 Figure 8. Effects of alternating white and UV sources on measured velocity. Purple color
25
26 corresponds to period of UV illumination, while orange corresponds to the period of white light
27
28
29 illumination.
30
31
32 Comparing the two colors in the plot shows that incident UV consistently induced an increase in
33
34
35
velocity. By contrast, replacing the UV source with the microscope’s white light source
36
37 generally brought a velocity reduction relative to the UV source. Thus, UV wavelengths greatly
38
39 enhance velocities, similar to the results shown in Figure 7.
40
41
42
43 The figure shows additional features worthy of mention. First, an overall falloff of velocity was
44
45 observed over time, similar to that seen without alternation. Second, close to the 20 minute mark
46
47 of each period of UV illumination, the velocity began to decrease. This decrease implies that the
48
49
50 enhancing effect of UV diminished over time, even with continued illumination.
51
52
53 Observing velocities under light illumination raised the question whether heat transfer might be
54
55
56
responsible for the increased velocity. Calculations show that the amount of energy coming from
57
58
59
60
14
1
2
3
the light leads to trivial temperature change: For the 2,000 µW light intensity (the maximum
4
5
6 used in these experiments), 0.35 mg water in the chamber, and specific heat of water of 1.0
7
8 Cal/g°C, calculations show 0.00008 °C temperature rise per hour. Thus, heating effects should be
9
10
11
inconsequential.
12
13
14
15
16
17 DISCUSSION
18
19
20
21 General Features of Flow
22
23
24 This paper reports the self generation of flow inside hydrophilic tubes immersed in water. The
25
26
27
flow phenomenon was first observed in this laboratory by Yu et al. (Yu, Carlson, & Pollack,
28
29 2013), who showed that the flow took place not only in Nafion tubes, but also in tunnels bored
30
31 through polyacrylic acid gels. We have confirmed their result, and extended the observations to
32
33
34 include flow in PEG and PVA channels. Thus, the flows are not specific to Nafion tubes, but
35
36 occur in a variety of tubes and channels made from hydrophilic materials. We have also
37
38 demonstrated that the energy for driving the flow comes from light.
39
40
41
42 Origin of Flow
43
44
45 The driving mechanism appears to come from light-induced separation of charge. Earlier studies
46
47
established a foundation for this view. When hydrophilic surfaces are immersed in water,
48
49
50 extensive solute-free zones form (Zheng et al, 2006). These “exclusion zones” or “EZs” bear
51
52 charge. Generally, the EZ is negatively charged, while the water beyond the EZ is positively
53
54
55 charged (Zheng & Pollack, 2006). Those studies also reported that the EZ’s negative electrical
56
57 potential has its maximum value close to the material surface, progressively decreasing farther
58
59
60
15
1
2
3
from that surface. Figure 9 (a) shows a schematic of this charge separation. Future studies should
4
5
6 be able to measure the electrical potentials in situ and quantify the UV-based increase in charge
7
8 separation.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24 (a) (b)
25
26
27
28 Figure 9. (a) Distribution of charge in the exclusion zone and water beyond the exclusion zone,
29
30 as shown by electrical potential measurements and pH sensitive dye studies. Protons spread in
31
32 bulk water, although some cling to the negatively charged EZ. (b) Disposition of charge
33
34
35 anticipated in tubular configuration.
36
37
38 Figure 9 (b) illustrates how a tubular geometry creates an annular EZ, resulting in a high
39
40
41
concentration of protons in the core of the tube. The EZ bears negative charge, while the bulk
42
43 water in the core is positively charged with protons. As the proton concentration builds following
44
45 the tube’s immersion in water, increasing repulsion causes protons to move axially towards less
46
47
48 positive areas outside the tube. This movement should result in a mass flow of water from inside
49
50 of the tube towards the outside.
51
52
53 Details of this process are summarized in Figure 10. As the exclusion zone builds, protons are
54
55
56 released to the bulk water (top). Concentrations of protons will be highest near the EZ-bulk water
57
58
59
60
16
1
2
3
boundary because of the protons’ attraction to the EZ’s negative charge. With continued input
4
5
6 radiant energy and further release of protons, increased repulsive force among protons drives
7
8 many protons toward the tube’s central axis (middle). Progressively, protons occupy the core
9
10
11
region, and the proton gradient from inside to outside exists throughout the entire tubular cross
12
13 section (bottom). Those protons near the central axis are not restrained by their attraction to the
14
15 EZ, and are therefore relatively freer to flow axially. They drive the flow.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53 Figure 10. Protons leave the EZ, initially concentrating at the EZ boundary (top). As more
54
55 charges separate, the increased numbers of protons, subject to proton-proton repulsion, begin
56
57
58
59
60
17
1
2
3
moving towards the central axis (middle). As the concentration of protons near the central axis
4
5
6 grows further (bottom), protons begin flowing out of the tube from one end or the other (bottom).
7
8
9 An expectation of this proton-based hypothesis is that protons will accumulate in the chamber.
10
11
12 Protons leaving the interior of the tube flow into the chamber continuously. Hence, the chamber-
13
14 water pH should progressively decline as flow continues. To check this expectation, we
15
16 measured the pH of the chamber water before and after the flow using a standard pH probe. The
17
18
19 pH decline was confirmed, initially at a rapid rate, which progressively decreased over time.
20
21 Over 2.5 hours, measurements showed a change from pH 7.48 to 2.77. When flow stopped, the
22
23 pH decline stopped. Thus, the anticipated release of protons in the flow-driving process is
24
25
26 confirmed.
27
28
29 Flow Direction
30
31
32
33
The flow began in a direction that was unpredictable. However, once it began, it continued in the
34
35 same direction until the flow finally stopped.
36
37
38 Probably the direction was set by some kind of asymmetry inside the tube. Proton concentration
39
40
41 is neither perfectly uniform along the tube axis nor symmetric at two ends, even though we tried
42
43 to use uniform tubes and adjusted the incident light to be reasonably uniform over the field. Any
44
45 axial asymmetry in proton concentration implies different inside-to-outside proton gradient at the
46
47
48 respective ends of the tube. One end’s gradient eventually overcomes that of the other and directs
49
50 the fluid out of the tube in the respective direction. Once that flow begins, fresh water is pulled
51
52
53
into the rear end of the tube, gets protonated, and the flow continues.
54
55
56
57
58
59
60
18
1
2
3
We considered several experimental approaches to try to purposefully create asymmetries. Such
4
5
6 experiments may seem simple and trivial, but several attempts achieved only limited success.
7
8 The only approach that proved successful was using tapered tubes. In that case the flow direction
9
10
11
was consistently toward the smaller end. Many variables can enter into the setting of direction.
12
13 Serious efforts in the future may settle this interesting but challenging question.
14
15
16 We speculate that those instances in which flow failed to start might correspond to instances
17
18
19 where the proton distribution remained relatively uniform along the tube. In those cases, similar
20
21 proton gradients exist at the two ends; hence one pull cancels the other. Often, in initially
22
23 “unsuccessful” experiments, we perturbed the system by mixing the fluid in the chamber,
24
25
26 presumably altering the local gradient. In most cases flow started. Furthermore, flow began at a
27
28 high success rate in the shorter tubes, where nonuniformities caused by “end effects” might be
29
30
31
anticipated to dominate. This suggests that in the very long tubes in which tube effects greatly
32
33 outweigh end effects, some initiator might be required to prime the flow.
34
35
36 The question might be raised whether local microsphere flow genuinely represents local flow of
37
38
39 the liquid. To be certain, we checked this by using pH dyes, which attach to hydronium ions
40
41 (protons plus water). Flow of pH dye (which turns red in the presence of protons) should
42
43 therefore be representative of the flow of water. We observed that the pH dye and microspheres
44
45
46 flowed together, at the same velocity. Hence, local microsphere flow velocity should be a valid
47
48 indicator of local water flow velocity.
49
50
51
52
Effects of Varying Parameters
53
54
55 Chamber Size
56
57
58
59
60
19
1
2
3
A key observation was that the flow lasted longer in larger chambers. In the largest chamber (6
4
5
6 ml) flow duration exceeded 1.5 days, with velocities above 15 µm/s over more than half the flow
7
8 time. That value is greater than the highest velocity recorded in a standard 350-µl chamber.
9
10
11
Decreasing the chamber size resulted in less total volume passing through the tube before flow
12
13 stopped (Figure 4).
14
15
16 These observations fit the proton-gradient hypothesis. The hypothesis posits that the inside-to-
17
18
19 outside proton gradient bears responsibility for driving the flow. Therefore, protons from inside
20
21 the tube flow into the water outside of the tube. Since larger chambers have more capacity for
22
23 accepting protons without reducing the chamber pH and thereby compromising the driving
24
25
26 gradient, those larger chambers should result in longer-lasting and more vigorous flow. That is
27
28 what was found and illustrated in Figure 4.
29
30
31
32
In smaller chambers, by contrast, protons accumulate more rapidly in the chamber.
33
34 Accumulation will diminish the proton gradient from inside to outside, progressively reducing
35
36 the driving force and compromising the flow. That was observed as well.
37
38
39
40 Tube Diameter and Length
41
42
43 Larger diameter tubes produced higher velocities. One contributing factor may be the increase of
44
45 tubular surface area, which should increase the rate of proton generation and thereby produce
46
47
48 faster flow.
49
50
51 On the other hand, increasing diameter also increases the fluid volume to be moved and the load
52
53
54
increases. Thus, the balance between increased drive and increased load needs to be considered.
55
56
57
58
59
60
20
1
2
3
This comparison is not entirely straightforward, however, because the driving force may vary
4
5
6 over the radial dimension: the circumferential protons tend to cling to the negatively charged EZ;
7
8 hence, they may not contribute much to the axial drive. The protons closer to the central axis
9
10
11
may be more effective. Large tubes should contain more of those central axis protons. On the
12
13 other hand, the exact radial distribution of those “central axis” driving protons is difficult to
14
15 predict quantitatively. Hence, the quantitative extent of velocity increase with increased tube
16
17
18 diameter is difficult to predict.
19
20
21 Regarding length, we found an optimum length that produced the fastest flow. For tubes longer
22
23 than that optimum, the flow diminished and sometimes did not even occur. In theory, the longer
24
25
26 tubes should generate more protons inside, hence higher drive, and that seemed to bear out at
27
28 lengths up to optimum. At the same time, the load increases proportionally with length due to a
29
30
31
larger volume of fluid to be moved. We think the balance between this two effects control the
32
33 optimum length for the flow at each diameter.
34
35
36 For the latter, however, a relevant factor may be proton concentration inside the tube. If protons
37
38
39 are not quickly swept out of the tube, as in the shorter tubes, then their lingering presence may
40
41 inhibit further release of protons, which would diminish flow.
42
43
44 Effect of Incident Light
45
46
47
48 Earlier observations on flow in Nafion tubes showed that the flow occurred spontaneously, i.e.,
49
50 solely by leaving the tube in the water (Yu, Carlson, & Pollack, 2013). Since the flow lasted for
51
52
53
extended periods of time, we wondered about the source of energy for driving the flow, and
54
55 tested whether the driver could be light. This idea arose from previous studies, which reported
56
57 that the EZ could be expanded by incident light (Chai, Yoo, & Pollack, 2009).
58
59
60
21
1
2
3
To carry out these studies, different light intensities were tested. At each optical power, the peak
4
5
6 velocity was measured. Higher light intensities resulted in larger peak velocities (Figure 6).
7
8
9 We also studied the effects of the energy-rich UV light in comparison with visible wavelengths.
10
11
12 At equal power, UV produced considerably higher velocities (Figure 7). Similar results were
13
14 found in experiments in which incident light alternated between visible and UV wavelengths
15
16 (Figure 8). As the presence of protons in the flowing fluid was confirmed by using pH dyes, the
17
18
19 faster flow under the UV light ought to be due to a higher rate of proton generation and charge
20
21 separation. Hence, UV wavelengths may be particularly effective in separating charge and
22
23 generating protons.
24
25
26
27 Changing light intensity and/or wavelength may therefore constitute practical means of
28
29 controlling the speed of the flow. This makes Nafion (and other hydrophilic) tubes versatile
30
31
32
fluidic pumps with no moving parts.
33
34
35 Conclusion
36
37
38 Nafion, and other hydrophilic tubes, were found to be self-priming fluidic pumps. The data
39
40
41 collected in these experiments support the hypothesis that tubular flow is driven by an axial
42
43 proton gradient between the inside and the outside of the tube. The effects of light intensity and
44
45 wavelength support this hypothesis: higher intensities of visible light as well as the energy-rich
46
47
48 UV light build larger proton gradients and thereby create faster flows. In other words, this “self-
49
50 driven” flow is driven ultimately by optical energy.
51
52
53
54
A fuller treatment of the role of optical energy on water may be seen in the recent book (Pollack,
55
56 2013).
57
58
59
60
22
1
2
3
4
5
6 REFERENCES
7
8 (1) Castellanos, A. Electrohydrodynamics. Springer-Verlag 1998.
9
10
11 (2) Davidson, P. A. An Introduction to Magnetohydrodynamics. Cambridge University Press
12
13
14 2001.
15
16
17 (3) Zheng, J. M.; Chin, W. C., Khijniak; E., Khijniak, E. Jr.; Pollack, G. H. Surfaces and
18
19 interfacial water: evidence that hydrophilic surfaces have long-range impact. Adv Colloid
20
21
22 Interface Sci., 2006, 127, 19-27.
23
24
25 (4) Szent-Gyorgyi, A. Bioenergetics. Academic Press 1957.
26
27
28 (5) Ling, G. N. In Search of the Physical Basis of Life. Plenum Press 1984.
29
30
31 (6) Goswami, S.; Klaus, S.; Benziger, J. Wetting and Absorption of Water Drops on Nafion
32
33
34 Films. Langmuir 2008, 24, 8627-8633.
35
36
37 (7) Chai, B.; Yoo, H.; Pollack, G. H. Effect of Radiant Energy on Near-Surface Water. J.
38
39 Phys. Chem. B, 2009, 113, 13953-13958.
40
41
42
43
(8) Yu, A.; Carlson, P.; Pollack, G. H. Unexpected Axial Flow through Hydrophilic Tubes:
44
45 Implications for Energetics of Water. Eur. Phys. Journal Special Topics, 2013, In Press.
46
47
48 (9) Zheng, J. M.; Pollack, G. H. Solute exclusion and potential distribution near hydrophilic
49
50
51
surfaces. Water and the Cell. Springer-Verlag 2006, 165-174.
52
53
54 (10) Pollack, G.H. The Fourth Phase of Water: Beyond Solid, Liquid, and Vapor.
55
56 Ebner and Sons, 2013.
57
58
59
60
23
1
2
3
4
5
6
7
8
9
10 TABLE OF CONTENTS ONLY
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
24

Flow Through Horizontal Tubes Submerged in Water

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Flow Through Horizontal Tubes Submerged in Water

Uploaded by

Copyright:

Available Formats

Subscriber access provided by BALL STATE UNIV

Peak Velocity [µm/s]

You might also like