who invented fiber optics

							(Funding source: NASA Grant NNC04GB52G, 2004-2009 and NSF Grant CTS-0086988, 2001-2005)

Invention Summary For: “Fluorescence and Fiber-Optics Based Real-Time Thickness Sensor for Dynamic Liquid Films”
Institution: Michigan Technological University Inventors: Amitabh Narain and Tian Ng (student) OUTLINE OF THE SENSOR’S FEATURES

The name of the invention
Fluorescence and Fiber-Optics Based Real-Time Thickness Sensor for Dynamic Liquid Films

What the invention does?
Non-intrusively measures time-varying local thickness of a dynamic liquid film.

How the idea of the invention was obtained or developed?
The basic idea for static films, but not this technology, has existed in the open literature. An earlier approach from the 1970s was upgraded by us and originally proposed to NSF with Dr. Anthony Smart (author of: Smart, A. E., and Ford, R. A. J., 1974, “Measurement of Thin Liquid Films by a Fluorescence Technique,” Wear, 29, pp. 41-47) as a consultant for the sensor development. The method was too unwieldy and could not address the issues and it was dropped. The reported invention and technology arose from the inventors’ desire to address all outstanding and unforeseen issues from the failed approach.

What problem or situation does the invention attempt to solve?
The invention provides a much needed instrument towards developing better understanding of wave-structure effects on transport processes (heat-flux, etc.) associated with dynamic films (film cooling, condensates, evaporating films, etc.). Potential users are university or industry sponsored experimental researchers (on film cooling, film condensation, film evaporation, annular two-phase flows with or without phase-change, etc.) who require time-varying thickness or wave-structure characterization of dynamic liquid films. Therefore the invention facilitates development of related cooling, pharmaceutical/chemical processes technologies.

What materials/tools were used to make the invention?
Tools employed are too many and too advanced to build it in house. The following assortment of vendors and tools/equipments are needed: a vendor that makes fiber-optic cables fitted to specified machined parts (distal tips, etc.), a vendor specializing in custom-made optics for properly collimating and focusing laser light into fibers and, also, collecting light from the fibers and collimating/focusing them on specified detectors, a vendor for custom-made optical filters, a vendor for non-fluorescent optical fibers, a suitable laser with laser drivers and temperature control (with the help of thermo-electric coolers) of the laser housing, fluorescence meter, spectrometer, spectrophotometer, a photomultiplier tube and its accessories, a sensitive photodiode and its accessories, etc.

What is the estimated cost for the materials to build the invention?
US $8,000 for single-probe sensor, US $10,000 for sensor with up to three probes

What is the estimated time it would take a person to build the invention?
For a custom order (inclusive of new liquid and new testing) from scratch, it may take 3-6 months depending on availability of a technical person doing the experiments. After all vendor supplied parts are available, assembly and testing may take two to three weeks.

What is the approximate total cost it would take to build the invention?
US $50,000 for the first few sensors for a new or different liquid. US $15,000 for building subsequent pieces of our FC-72 sensor

How much money could the invention be sold for?
US $55,000 for the first few entirely new designs and US $20,000 or so for building, testing, and calibrating subsequent pieces.

What would the profit be on each invention if it could be sold?
About US $5,000

How the invention was tested?
By direct comparisons of sensor results with results obtained from a separate experiment which allows time-varying thickness to be measured by an independent intrusive mechanical needle.

What the results of the testing or trials were?
The results established the sensor’s excellent ability to resolve the amplitude and frequency content of the dynamic film.

What improvements on the invention might be made to improve it?
The sensor’s resolution and range can be improved. The sensor validation experiment can be made more elaborate to check thickness by an alternative means at each instant of time.

What are the requirements for this sensor to work?
The film thickness sensor that has been invented is for measuring thickness of dynamic films formed by the liquid called FC-72 (perfluorohexane or C6F14), a fluid made by 3M Corporation. However custom-built technology for other liquid films can be made available whenever the following requirements are met: i. The liquid film (whose thickness is to be measured), if not naturally fluorescent over a convenient spectrum, can be doped by a suitable fluorescent chemical. If doping is not possible in the real process, it is assumed that it is allowed in a prototypical process. ii. The measurement must take place in a dark environment with no ambient light or only under special ambient light illumination over pre-selected and allowed wavelengths. iii. It should be possible to ensure that none or negligible fluorescent light is induced when the excitation light interacts with the material(s) forming the environment surrounding the film. iv. The liquid film thickness should be within a certain well defined range (which is 0.5 mm to 3.0 mm for the reported FC-72 films). A wider range of measurements and improved resolution (from the current ± 0.1 mm to ± 0.05 mm) is possible by suitable sensor modifications.

OBJECTIVE
To report design, development, and calibration of the invented sensor. The sensor utilizes principles of fiber-optics and fluorescence and measures a dynamic liquid film’s time varying thickness between 0.5 mm to 3.0 mm.

PRINCIPLE/FUNCTION
The sensor accurately measures the instantaneous thickness of a dynamically changing liquid film in such a way that the probe does not perturb the flow dynamics in the proximity of the probe’s tip. This is achieved by having the probe’s exposed surface embedded flush with the surface over which the liquid film flows, and by making arrangements for processing the signals associated with the emission and collection of light (in distinctly different wavelength windows) at the probe’s flush surface. A film thickness in the range of 0.5 to 3.0 mm can accurately (with a resolution that is within +/- 0.09 mm over 0.5 to 1.5 mm range and within +/- 0.18 mm over 1.5 to 3.0 mm range) be measured by the available sensor. Although this demonstrates the sensor’s

ability for dynamic film thickness measurements carried out for a doped liquid called FC-72 (perfluorohexane or C6F14 from 3M Corporation), the approach and development/calibration procedure described here can be extended, under similar circumstances, to some other liquid films as well. To fluoresce the liquid film, the FC-72 is slightly doped (concentration = mass of dopant * 100 / mass of total solution = 0.20%) by a suitably chosen fluorescent additive (bi-acetyl or C4H6O2 in this case). The excitation of this doped liquid film at shorter wavelengths of light (400-410 nm) leads to sufficient absorption of this light. This, in turn, induces sufficient fluorescent emission at longer wavelengths (460-600 nm). In other words, molecular absorption of shorter-wavelength photons triggers emission of lower-energy photons with longer wavelengths. The wavelength spectrum of emitted fluorescent light (460-600 nm) is separated and identified for detection because light in this spectrum - as it traverses its way out of this film - is not self-quenched, or absorbed by the doped liquid film. The excitation light is arranged to be minuscule in order to excite only a primary illumination cone (i.e., to minimize presence of secondary illumination cones resulting from interfacial reflection of the excitation light in the primary illumination cone) and, also, to avoid crossing a certain threshold level of detected fluorescent light that actually arrives at the sensor’s axially symmetric collection area indirectly after its reflection from the interface of the liquid film. As a result, the detected light (after filtering out the excitation light) in the fluorescent range is also minuscule and is detected by a sensitive photo-multiplier tube. The predominant amount of detected light comes directly from the fluorescent light originating in the primary illumination cone, and a much smaller amount comes indirectly – after its interfacial reflection. This predominant component of detected light correlates directly with instantaneous film thickness and is actually made precisely known (and deterministic) - by a process of sorting and identifying this portion of the total detected light with the help of a paired set of calibration experiments especially developed for this purpose. The results of the calibration experiments, as well as experimental demonstration of the sensor’s dynamic measurement capabilities, are shown below.
The amount of “green” fluorescent light collected from the fluorescent material depends only on thickness d for a given concentration of the dopant . 200 µm fiber-optic cable Violet Laser Light Source Film Thickness Sensor Probe Greenish fluorescent light is collected in the outer cable Violet illuminating light Liquid film d of 0.5 to 3.0mm

d

1 mm-dia fiberbundle cable

Suitable Long Pass Filtering Fluorescent Signal Detector (Photomultiplier Tube)

(B)

(A) Fig. 1: Operating principle (A) and an enlarged view (B) of the optical fiber configurations in the sensing plane of the “film thickness sensor” probe.

(A)

(B)
Fig. 2: A schematic design (A) and an associated hardware (B) for liquid thickness sensing system.

(B) (A)

Figs. 3: The proposed and implemented “paired” set of calibration experiments (A: with L-V interface; B: without L-V interface).

NOMENCLATURE AND REQUIREMENTS
NP (mV): Noise light received through R despite all the filtering at I-tip, R-tip, and elsewhere. Sf|L-V (mV): Fluorescent signal for a doped film thickness (with L-V interface, Fig. 3A) D (mV): Detected light for a doped liquid film, D = Sf|L-V + NP NP/D (dimensionless): Noise-to-detected light ratio (also called noise ratio). Sf|No L-V (mV): Ideal fluorescent signal for a doped film thickness (w/o L-V interface, Fig. 3B) Design Goal: D ≈ Sf|L-V ≈ Sf|No L-V so the desired “one-to-one” correspondence can be achieved (as shown below)

• •
• •

• •

Optimum setting (with 6 receiving fibers), 40.8 mA, Aug 9, 2006
250.0
180.0 160.0

Optimum setting (with 6 receiving fibers), 40.8 mA, Aug 9, 2006

200.0

140.0 120.0 S f|L -V (mV)

D (mV)

150.0

100.0 80.0 60.0

100.0

50.0

40.0 20.0

0.0 0.0 20.0 40.0 60.0 80.0 100.0 120.0 140.0 160.0 180.0 S f|L-V (mV)

0.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 110.0 S f|No L-V (mV)

Optimum setting (with 6 receiving fibers), 40.8 mA, Aug 9, 2006 y = 37.538x - 9.9244
R2 = 0.9979 110.0 100.0 90.0 80.0 (m V )
L -V

70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Film Thickness (mm)

Figs. 4: Calibration achieves one-to-one correspondence that does not require total elimination of NP or significant enhancement of D/NP or Sf|L-V/NP.

S f|N o

SENSOR’S DYNAMIC RESPONSE VERIFICATION

(A)

(B)

Figs. 5: In (A), dynamic signal D (in mV) is for a dynamic film whose “local” timevarying thickness at a point is being sought. In (B), the sensed signal D has been transformed, with the help of calibration results in Fig. 4, to yield local time-varying film thicknesses (in mm) measured by the sensor at a fixed location. The frequency and amplitude of the film thickness in (B) are in excellent agreement with alternative and independent (mechanical) measurements (marked x in B).

CONCLUSIONS
1. Invention and calibration of a sensor for dynamic liquid film thickness measurements has been completed. 2. D and Sf|L-V increase with increasing liquid film thickness. 3. Sf|L-V/NP has been increased to at least 2 to 11 for d = 0.5-3.0 mm. 4. Liquid interface configuration independence for signals D and noise NP has been achieved. 5. Total errors associated with D can be minimized to within +/5.1%. A thickness resolution of +/- 0.09 mm (0.5-1.5 mm) and +/0.18 mm (1.5-3.0 mm) has been achieved and can be further improved.

REFERENCE
Ng, T. W., 2006, “Development and Calibration of a Fluorescence and Fiber-Optics Based Real-Time Thickness Sensor for Dynamic Liquid Films,” PhD Dissertation, Michigan Technological University.