FAQs
1. What is the difference between a reflectometer and an interferometer?
2. Which do ORS make?
3. How do ORS’ reflectometers work?
4. What makes ORS solutions better than their competitors?
5. What is the difference between an EpiEYE and a MiniEYE family of products?
6. Why are there two family types?
7. What if I don’t have any windows on my reactor?
8. Can I install the instrument myself?
9. Is it easy to set-up?
10. Is there a warranty?
11. What does it cover?
12. Can I extend it?
13. How long will the laser last?
14. How many lasers are in the MiniEYE/EpiEYE?
15. What wavelength are the lasers at?
16. Why are these standard wavelengths?
17. How does the pyrometer work?
18. What is emissivity?
19. Why is it important to correct it?
20. What is the temperature range of the pyrometer?
21. Can it go lower/higher? Why?
22. What is the minimum radius of curvature of your wafer bow?
23. Why do I need in situ monitoring?
24. Why hasn’t everybody already got one?
25. What are the differences between MOCVD and MBE?
26. Do I need a different instrument for each type and if so why?
27. I grow on a transparent material. Will it still work?
28. I grow while rotating very slowly, typically at 3 rpm. Will it still work?
29. I grow while rotating very quickly, typically at 1000 rpm. Will it still work?
30. My growth rate is 500nm per second. Will it work?
31. I grow the weirdest material in the world. Will it still work?
32. I grow homo-epitaxial materials (GaAs on GaAs). Will it still work?
33. Do you have any data for VCSEL/whatever structures?
34. Why don’t you use white light and filter out the wavelength you are interested in?
35. Will the laser affect the growth?
36. Over what area can it see the growth?
37. If the spot was bigger would it not give me more data?
38. Does ambient light affect the reading for growth/pyrometry?
39. How do you overcome this?
40. How do I know it is always measuring the same point?
41. What is the difference between R-fit v4 and R-Fit LIVE?
42. What are the benefits of R-Fit LIVE over R-Fit v4?
43. How live is R-Fit LIVE?
44. How will R-Fit LIVE help me make more wafers?
45. What information does R-fit give me?
46. What is better about R-Fit to competitors’ software?
47. Will R-Fit work with other reflectometers?
48. How many licences can I have?
49. Will it export to word/excel?
50. Do I get free upgrades to the software?
51. What is the lead time for an instrument & software?
52. Can I have a discount for multiple sales?
53. What support do I get?
54. Is installation included?
1. What is the difference between a reflectometer and an interferometer?
An interferometer is a device for measuring very small changes in thickness, a reflectometer measures changes in reflectance. We sell a reflectometer which measures changes in reflected light intensity, the changes are due to the thin film causing interference fringes; the thin film is the interferometer.
The Reflectometer.
In its simplest form a reflectometer consists of a laser, a beam-splitter and a photo-detector arranged as shown below;
Provided the thin film that is being deposited is transparent at the wavelength of the laser, then the incident laser beam will undergo multiple reflections at the thin film / substrate interface and the thin film surface as shown below;
In effect, the thin film is itself acting as an interferometer. These reflected rays will interfere constructively (bright fringe) or destructively (dim fringe) depending on the thickness d(t) of the thin film.
Reflectometers, pyrometers and analysis software. We also make wafer bow equipment and post growth wafer imaging systems.
3. How do ORS’ reflectometers work?
Basic operation is described in answer 1; laser mounted above a cube beam splitter with a photo-detector to one side. Laser beam passes through cube, hits target, is reflected back into cube which directs beam into photo-detector. Photo-detector converts light to a current which is recorded.
4. What makes ORS solutions better than their competitors?
We have gone to a lot of trouble to make ours stable, we are able to trigger ours at a set time or angle relative to a homing pulse and we can make multiple measurements per rotation in a rotating (multi-wafer) reactor. Ours is also automatically back-ground corrected so the measured signal is ‘absolute intensity’ allowing for real-time fitting of the data. Ours converts data into information straight away. Ours can be fitted onto any make or model of reactor. We have a very large optical access so even if the return reflected beam does not share the same optical path as the transmitted beam, we can still capture it; particularly useful in MBE reactors. Also, because we can synchronise the capture of the beam to the rotation of the platen, we can all but eliminate intensity variations due to wafer wobble (precession; something 95% of reactors suffer from, particularly MBE). It is better because we designed it that way in response to the very large number of customers we have dealt with over the years, all with different reactor platforms.
5. What is the difference between an EpiEYE and a MiniEYE family of products?
The EpiEYE has the laser and lots of the (sensitive) control electronics in one optical head. The MiniEYE does not have the laser in the head, almost all of the electronics is in the 19” rack; it can take up to 3 wavelengths and it is small and light weight in comparison to the EpiEYE.
6. Why are there two family types?
Because there are so many different types of reactors out there; the MiniEYE was originally designed and built to fit onto ‘flip-lid’ reactors where size and weight might be a problem; the EpiEYE has found a good niche with applications where precession is an issue since it is the model with the largest optical access (circa 2.5cm compared to 1cm with the MiniEYE). The EpiEYE family also contains two other variants, the UniEYE and the AccuMET.
7. What if I don’t have any windows on my reactor?
Then you need to get some – if you can’t get light in then you can’t get it out and there’s nothing to see.
8. Can I install the instrument myself?
Yes.
Yes – A full installation manual will be provided.
Yes – 12 months return to base warranty.
Failure of components not due to damage from misuse.
Yes – Contact ORS Ltd for details.
13. How long will the laser last?
Generally -10,000 hours.
14. How many lasers are in the MiniEYE/EpiEYE?
This depends on the model;
UniEYE, EpiEYE Basic and MiniEYE Basic have one each.
AccuMET, EpiEYE Standard and MiniEYE Standard have two.
It is also possible to have custom instruments from either family that have three lasers.
15. What wavelength are the lasers at?
Typically;
UV – 375nm
Blue – 405nm, 445nm, 473nm,
Red – 635nm, 650nm, 670nm,
Green – 532nm
IR – 780nm, 805nm, 850nm, 905nm, 950nm, 980nm, 1064nm, 1310nm
and 1550nm.
16. Why are these standard wavelengths?
The red is convenient – you can see it so lining up is a lot easier. The IR is needed for pyrometry. The blue is required in some instances for monitoring from nitrides and when used in combination, it is possible to compute composition.
17. How does the pyrometer work?
It is basically a photo-diode that just looks for light that any hot body gives off (as in wow! that poker is red hot!). The hotter an object, the more light it gives off and it is this that a pyrometer measures. In the instrument, you have to select a narrow wavelength range at which to make the measurement.
A property of all materials; it determines how efficient a surface is at allowing light to pass through.
19. Why is it important to correct it?
The theory used for converting the measured light from a hot object into a temperature assumes that the objects surface is a ‘perfect transmitter of light’; the so called ‘black body’. No bodies are black bodies so we have to measure the emissivity and correct for it. Furthermore, as a thin film is being deposited, the emerging light will be subject to interferometry fringes. This is why its important to use a laser or diode to measure the fringes and then correct for them in the pyrometry, hence the name; Emissivity Corrected Pyrometry (ECP).
20. What is the temperature range of the pyrometer?
Circa 450°C to 1500°C
21. Can it go lower/higher? Why?
It can go as low as 400°C but no lower. This is because ‘warm’ objects don’t give off enough light for the photo-detector to detect. It can go up to very high temperatures; the problem then becomes saturating the photo-detector; it has a finite range and as bodies get hotter they give off more and more light.
22. What is the minimum radius of curvature of your wafer bow?
100m – Changes of around a micron can be seen.
23. Why do I need in situ monitoring?
If you want to know;
• What you’ve deposited.
• How much you’ve deposited.
• What is the surface quality?
• How fast you are depositing the layer?
• The uniformity across an individual wafer.
• The uniformity wafer to wafer.
• The composition of the layer.
• The temperature of the substrate and layer.
• The degree of bend induced in the wafer during growth.
It is also important to be able to perform accurate statistical process controls on what has been grown.
24. Why hasn’t everybody already got one?
Several reasons;
• They don’t have optical access
• They have got along so far without it since the specifications they are working too we slack enough to allow them proceed.
• Culturally they don’t understand what it’ll do for them.
• It’s perceived as a nice to have and not an essential (this is true for some materials but for nitrides and others this is certainly not the case)
• It is perceived to be expensive: However, the return of investment is typically less than 6 months if used correctly
25. What are the differences between MOCVD and MBE?
In essence, the various techniques are the same and involve the controlled deposition of tiny particles onto the surface in question as depicted in the above diagram. There are several methods that have evolved over the years:
a) Molecular Beam Epitaxy (MBE) in which the particles are atoms or molecules e.g. As2 or Si evaporated from the solid under ultra-high vacuum conditions.
b) Metal Organic Vapour Phase Epitaxy (MOVPE, also known as Metal Organic Chemical Vapour Deposition MOCVD) where the particles are organic molecules containing a metal atom. The molecule ‘cracks’ on the surface to leave the metal atom behind, the organic part being volatile moves away from the surface.
c) Ion Beam Sputtering involves nano / micron size globules of the element being formed and directed at the surface under high vacuum condition.
Irrespective of the deposition technique, it is possible to monitor the development of the thin film as it is being deposited using optical techniques.
26. Do I need a different instrument for each type and if so why?
The technique doesn’t matter, but the geometry does and it is this that determines what type of optical head you need.
27. I grow on a transparent material. Will it still work?
Yes, although you have to be careful if your substrates are polished on both sides as this can lead to unwanted thermal fringes.
28. I grow while rotating very slowly, typically at 3 rpm. Will it still work?
Yes.
29. I grow while rotating very quickly, typically at 1000 rpm. Will it still work?
Yes.
30. My growth rate is 500nm per second. Will it work?
It depends on other factors too; if you are growing on a single wafer which is under the EpiEYE at all times, then yes; if you are growing on a multi-wafer system with the wafers passing under the EpiEYE then it depends on the rotation speed; if the speed is slow, then no, if it is fast then yes. If it is slow then you just won’t be able to record enough data.
31. I grow the weirdest material in the world. Will it still work?
Provided the material is transparent to al least one of the laser wavelengths we use, then yes. As an example, it is possible to monitor thin layers of gold or other metals (thin is less than about 200-300Å).
32. I grow homo-epitaxial materials (GaAs on GaAs). Will it still work?
No – As there is no difference in refractive index between the substrate and the deposited material - unless you dope your materials
33. Do you have any data for VCSEL/whatever structures?
Yes – Lots - See the technical notes.
34. Why don’t you use white light and filter out the wavelength you are interested in?
The intensity will be very low compared to a laser. This means the noise will be high and so there will be considerable uncertainty in the results. It is also more difficult to focus a white light beam (so a problem for MBE) and more difficult to control the stability of the light (so noise again becomes an issue). Extracting accurate reflectance data at sufficiently narrow wavelength bandwidth can also be a problem, leading to errors in the analysis of growth rate and thickness.
35. Will the laser affect the growth?
Not normally due to the low intensity of our laser diodes and the fact that they are modulated so spend some of their time off. This can be an issue with some II-VI materials but not for III-V and group IV materials.
36. Over what area can it see the growth?
Over the area of the laser beam, roughly a circle of diameter about 2mm (so roughly 3 – 4 mm²) – It is possible to profile across a wafer so you are able to see more area if required.
37. If the spot was bigger would it not give me more data?
No, it would give you more averaged data; not necessarily a good thing. The spot size is keep small to allow the beam to pass into the reactor and come out again without hitting something you don’t want it to.
38. Does ambient light affect the reading for growth/pyrometry?
Yes to both.
For reflectometry this is not a problem unless the heating is via IR lamps (even then we can introduce a narrow band pass filter to remove its effects). Our emissivity corrected pyrometry has been designed with special optics to be insensitive to such thing as Knudsen cells.
40. How do I know it is always measuring the same point?
The instrument is designed to make a reading at a specific time after it senses the homing pulse. Therefore, you are sure to be hitting the same spot at each rotation of your susceptor.
41. What is the difference between R-fit v4 and R-Fit LIVE?
R-Fit LIVE requires the refractive index to be input, then it automatically fits the incoming data to provide rate and roughness. R-Fit v4 cannot provide data in real time, but does not need an accurate value of refractive index to be put in. Both have to have ranges of values for the rate and the roughness in order to work.
42. What are the benefits of R-Fit LIVE over R-Fit v4?
Real time, instant results and information on which you can act. It will also give you information on which wafers in a multi-wafer reactor are growing on spec and which are not; this can be due to simple things like the wafer has not seated correctly in its pocket. R-Fit LIVE is for production whereas R-Fit v4 covers both production and research.
Within 0.25s of the layer going down you have the information.
44. How will R-Fit LIVE help me make more wafers?
It helps you improve your efficiency by flagging up any problem you are having very early on in the growth. It allows you to calibrate your reactor immediately before you start the structure thereby allowing you to correct the recipe and hence more of your wafers will be within specification. Because R-Fit LIVE provides real time information, then it can be used for closed-loop feedback control.
45. What information does R-fit give me?
Both packages give you information on;
• Growth rate
• Thickness
• Surface roughness
• Run to run reproducibility
• R-Fit v4.0 also gives you information on;
• Refractive index.
46. What is better about R-Fit to competitors’ software?
We use a ‘matrix’ approach to the maths which has no approximations in it. They use the virtual interface approach. They use look up tables, we don’t, we calculate everything. We can cope with any number of layers of any thickness (from about 20 Å) upwards.
47. Will R-Fit work with other reflectometers?
Yes – Subject to correct output of data.
48. How many licences can I have?
As many as you like.
49. Will it export to word/excel?
Excel/Access/any CSV reader.
50. Do I get free upgrades to the software?
No – Any bugs or fixes will be provided free of charge – Upgrades will be discounted if you are an existing customer.
51. What is the lead time for an instrument & software?
Software is a week. Hardware is typically between 4 and 12 weeks depending on the solution required.
52. Can I have a discount for multiple sales?
Yes – Discounts are available for multiple purchases.
30 days unlimited telephone and email support – Other support packages are available. Contact ORS Ltd for a quote.
No – Installation is designed to be simple. However, we are happy to install if you need our support. Contact ORS Ltd for a quote.
.......................................
- Latest news
Y-Systems chooses ORS Ltd- Vigo Systems get great results from ORS II/VI solution
- Worldwide agencies:
ORS anounce worldwide agency network
- Conferences
- Semicon Taiwan 2010, September 8-10 2010, Taipei World Trade Center, Taipei, Taiwan
- You're invited to visit Optical Reference Systems at Semicon Taiwan 2010 at the Taipei World Trade Center in Taipei, Taiwan. We are located in the South Hall at Booth # 2929
- Click Here for the Floorplan
- Latest products
- EpiREF: Easy to install , laser based in situ reflectance monitor
- EpiEYE: Combines the laser based reflectance of the EpiREF with temperature measurement
- EpiEYE Blue: All the features of the EpiEYE with additional measurements at 405nm
- ORS ARC: A single beam in situ wafer bow measurement tool
- R-Fit Live: Increases reactor up-time and wafer yield
- Get Adobe Reader here
E-mail: enquiry@ors-ltd.com