Chapter 8.34

Nanometrics 210 XP Scanning UV

Nanospec/DUV Microspectrophotometer

(NanoDUV)

1.0 Title

Nanospec/DUV Microspectrophotometer (NanoDUV)

2.0 Purpose

The Nanospec®/210 XP Scanning UV is a computerized film thickness measurement system. It is an old generation of NanoSpec (Chapter 8.33). The monochromator is mechanical and the software runs on DOS environment. However, there are some features that NanoSpec does not provide.

► Deep UV light source that extends the measurement limit down to 100Å.

► The refractive index can be entered by user as an option.

► Capable of measuring underlayer film thickness.

► Wafer stage that accommodates 6” wafer.

3.0 Scope

This manual covers the operating procedure of film thickness measurement, system re-initialization, and user level trouble-shooting guide. The available measurement programs for various film stacks are listed in Appendix 12.1.

4.0 Applicable Documents

4.1 Nanometrics AFT Operations Manual (copy in Office)

4.2 NanoSpec Film Thickness Measurement System, Microlab Online Manual Chapter 8.33.

5.0 Definitions & Process Terminology

N/A

6.0 Safety

6.1 Burn Hazard: The light source is hot. Be careful when aligning the light bulb.

6.2 Strong Light Hazard: Do not look into the light source directly for a long period of time.

6.3 UV Light Hazard: Do not look into the UV light source.

7.0 Statistical/Process Data

N/A

8.0 Available Processes, Gases Process Notes

8.1 Standard programs provided by the instrument vendor are listed in Appendix 12.1.

8.2 In order to measure the film thickness, the refractive index of the film is needed. A user can use the default value stored in the computer or enter it before the measurement.

8.3 NanoDUV can only measure transparent film. Metal film and thick Germanium/Silicon film can not be measured.

9.0 Equipment Operation

System Description

9.1 The Nanospec system consists of a customized optical microscope and a personal computer with DOS OS. The optical microscope has a mechanical spectrophotometer head and a dual wafer stage, left side for the reference wafer and right side for the sample to be measured. The microscope has three objective lenses: 5X, 10X, 15X (for UV measurement ONLY), and 50X magnifications.

9.2 All the system operations are controlled by the specially labeled keys on the computer keyboard. The functions of these keys are as follows:

CALIB Stops any measurement in progress and returns to the Available Programs screen. This gives the operator an opportunity to choose another program, objective lens or to run a new reference scan.

NEW TEST Initiates a new measurement sequence with a new sample using a previously selected film program. A zero check is generally made and new sample identification, which can be blank, must be entered.

ENTER Terminates an alphanumeric entry. The computer has not received the data until <Enter> is pressed.

MEAS Initiates a measurement. It can also be used to respond to a prompt requiring a "YES" answer. Holding down the <MEAS> key causes the system to make continue measurements.

YES/0 Is used to respond "YES" to certain prompts ending with a question mark, or as the zero key in a numerical entry.

NO/+ Is used to respond "NO" to certain prompts ending with a question mark, or as a decimal point in a numerical entry.

HELP Explains what the commands are and assist you answering the questions.

R/I ON/OFF Toggles on and off the instrument's option to measure the refractive index of a single layer film. The Refraction Index Option must have been enabled (Section 9.9.3) during the initialization for this function to work.

DUAL Measures the thickness of both layers of films with a single measurement. The estimated underlayer thickness needs to be entered for each sample.

GRAPH Displays an interference spectrum (Reflectance versus Wavelength) for both the measured sample and the mathematical model corresponding to the measured thickness. When pressed, the screen shows the previous stored spectrum. Follow the computer instruction to clear it and display the current measurement.

DISPLAY

STATISTICS is used to display statistics for measurements taken on an individual sample during a measurement session.

Measurement Procedures

9.3 Enable NanoSpec on the WAND. Turn on the LCD screen and press any key to stop the screen saver program. Make sure the light source is on. If there is no light in the microscope, report on the WAND.

9.4 If need UV light for thin film measurement (program 17 to 20), turn on the UV light source on the wall shelf to the left of the system. Wait at least 20 minutes for the UV light source to stabilize.

9.5 Place the reference wafer (clean bare silicon wafer) on the left side and the sample to be measured on the right side of the stage.

9.6 Press the CALIB key to display all available programs on the screen. Select the program for the measurement by entering the program number. Enter the objective lens magnification, and refractive index when prompted. To use the default refractive index, just press Enter key.

9.7 The computer will prompt you to do a reference calibration. Follow the screen instructions.

9.7.1 Depends on the program used, it may ask to open or close UV light shutter. The shutter is located on the black box behind the wafer stage.

9.7.2 Use the tab at the front center of the wafer stage to move the stage to the right end, so that the reference wafer is under the objective lens.

9.7.3 Use the coarse (large) and fine (small) knobs at the base of the microscope to focus on the reference wafer. Since there is no pattern on the reference wafer, focus on the octagon in the field of view.

9.8 After the reference calibration, the tool is ready to make measurements.

9.8.1 Use the tab on the front center of the wafer stage to move the stage to the left end, so that the sample is under the objective lens.

9.8.2 Focus on the sample. Move the wafer stage using the X-Y axis knobs under the stage.

9.8.3 When measuring patterned sample, make sure the black dot, which marks the active optical area, in the center of the objective lens is well within the center of the patterned feature to be measured. The actual dot size depends on the magnification of the objective lens.

Objective Magnification	Meas. Dot Size (μm)
5X	30
10X	15
15X (UV only)	10
50X	3

If you need to change the objective magnification now, you have to redo the reference calibration again. Use the same magnification for the calibration as for the sample measurement.

9.8.4 Press the MEAS key to make a measurement. To ensure the sanity of the measurement, you should press the GRAPH key to see how the measurement algorithm fit with the actual spectrum. The rule of thumb is when the peaks and valleys of the spectrum fit well, the measurement should be correct.

9.8.5 Write down the data since there is no printer attached to the tool. Disable the tool on the WAND when done.

System Initialization

9.9 The NanoDUV needs be initialized only when it does not work properly, or the user wants to change the Refractive Index Option. To start the initialization procedure, hold down the control key, and type r-e-s-e-t (reset).

9.9.1 A startup message will appear on the screen, along with the question: Enable Datalink? Answer <NO> because this function is no longer supported.

9.9.2 The next prompt asks “Is WAVELENGTH 480?” Check the wavelength displayed on the spectrophotometer head. If the wavelength is 480, simply press the <YES/0> key. Otherwise, press <NO/+> and then enter the current wavelength shown on the display (3 digits only) at the "Enter Wavelength" prompt. The instrument will adjust the wavelength, and prompt you again. Another repetition of this process may necessary to set the wavelength at 480 nm. Once the wavelength is 480 and you have answered "YES" to the prompt, the question

9.9.3 The next prompt asks “Refractive Index Option?” Answer <YES> to enable this feature, giving users the option of entering a refractive index for a given film other than the default. Answer to disable the R/I ON/OFF key and use the system default values.

9.9.4 The last question is “Enable Printer?” Answer <NO> since there is no printer connected.

9.10 Then follow the screen instructions to adjust the Gain controls. The Zero and Gain controls are both located inside the cover under the spectrophotometer head. Adjust these by inserting a forefinger in the open left end of the cover. The first knob (small metallic) is for the Zero and the second (bigger plastic) is for the Gain.

9.11 When initialization is complete, the list of available programs will appear on the screen. The system is ready for measurement.

Special Instructions for UV Programs

9.12 Turn on the DUV lamp power supply. It locates on the back shelf of the tool. It needs 20 to 30 minutes to stabilize. You can feel by touch that the lamp case is warm afterwards. If not, the lamp is not working and needs to be replaced.

9.13 The native oxide on the reference wafer will affect your measurements. For consistent results, you should dip the reference wafer in HF to get rid of the native oxide. Follow the sink operation procedures to rinse and dry the reference wafer before put it on the stage.

9.14 Use only 15X (UV) objective lens to do your measurements.

10.0 Troubleshooting Guidelines

10.1 Problem: Computer does not respond to the keyboard.

Cause: Computer OS/hardware problem.

Solution: Reset the computer by switching it off/on.

10.2 Problem: The measurement spectrum does not agree with the computer algorithm.

Cause: The reference calibration was not done correctly.

Solution: Press the Calibrate key and re-start the calibration.

10.3 Problem: No light in the microscope

Cause: Light bulb burned down or power source turned off.

Solution: Report the problem on the WAND.

11.0 Figures & Schematics

N/A

12.0 Appendices

12.1 Program Descriptions

The system holds 20 predefined standard programs, and 32 user-defined programs:

Prg. #

Type of Film

Standard R.I.

Wavelength Range (nm)

252

253

254

255

Oxide on Silicon

Nitride on Silicon

Negative Resist on Silicon

Polysilicon on Oxide

Negative Resist on Oxide

Nitride on Oxide

Thin Oxide on Silicon

Thin Nitride on Silicon

Polyimide on Silicon

Positive Resist on Silicon

Positive Resist on Oxide

Reflectance Mode

Thick Films

EPA914EZ Resist on Silicon

Very Thin Oxide on Silicon

Very Thin Nitride on Silicon

UV Thin Oxide on Silicon

UV Thin Nitride on Silicon

UV Oxide on Polysilicon

UV Oxide on Aluminum

OxiNitride on Silicon

OxiNitride on Oxide

Special Substrate

Non-Standard Dual Layer

Single Layer Films

Special Program

1.456

2.001

1.550

3.973

1.550

2.001

1.456

2.001

1.781

1.631

1.600

1.631

1.447

1.932

1.447

1.770

480 – 800

520 only

480 – 800

Choose 1-3 wavelengths

720- -760

480 – 800

Not Available

200 – 280

220 – 280

210 – 280

480 – 800

1. Oxide on Silicon

This program measures silicon dioxide on silicon prepared by thermal oxidation. Measuring highly doped silicon dioxide may affect accuracy, because heavy doping changes the film's optical properties. The instrument will display "Under-range" when measuring silicon dioxide less than 100 A thick.

2. Nitride on Silicon

Use for measuring silicon nitride that has been chemically deposited on silicon. Measuring plasma nitrides may be less accurate than measuring chemically deposited nitrides because the varying stoichiometry of plasma films can cause measurement error. Frequently, the refractive index (if known) can be used to scale results obtained from plasma nitride samples. For films under 500 Å, use program 8. The instrument will display "Under-range" when measuring silicon nitride of less than 100 Å.

3. Negative Resist on Silicon

Measurements made with this program are in the range 500 - 40,000Å. The refractive index (RI) for the particular negative resist must be known. The default RI is 1.55.

4. Undoped Polysilicon on Oxide

Undoped polysilicon in the range of 550 - 10,000 Å can be measured using this program. When used on polysilicon films less than about 300 Å the message "under-range" is displayed. Research indicates that measurements of thick polysilicon (approximately 3,000 - 10,000 Å) are highly accurate when the range of underlying silicon dioxide thickness is between 300 - 1,200 Å. Measurement for thin polysilicon (less than 3,000 Å) is most accurate when the underlayer thickness is in the range of 800 - 1,200 Å. The instrument will display "Under-range" when measuring oxide of less than 10 Å.

5. Negative Resist on Oxide

A resist film between 4,000 - 30,000Å on an oxide film between 500 - 15,000 Å can be measured with this program. The program contains an extra prompt for the operator requesting that the oxide thickness be typed in. The refractive index (RI) for the particular negative resist must be known. The default RI is 1.55. It is important to enter an accurate, premeasured oxide thickness. Any error entered in the thickness results in an error in resist thickness readout. The instrument will display "Under-range" when measuring oxide of less than 10 Å.

6. Nitride on Oxide

This program measures the thickness of chemically deposited silicon nitride. The silicon nitride thickness must be greater than 100 Å. The instrument will display "Under-range" when measuring silicon oxide of less than 10 Å.

7. Thin Oxide on Silicon

This program averages data at 520 nm rather than scanning through a wavelength range. A new reference should be taken whenever selecting this program. When measuring very thin films, take a new reference every 30 minutes or more often. For increased confidence in very thin film measurements, repeat measurements. The instrument will display "Under-range" when measuring thin oxide of less than 20 Å.

8. Thin Nitride on Silicon

Programs 7 and 8 average data at one wavelength, 520 nm, for about 3 seconds, instead of scanning through the visible wavelength range (480 - 800 nm). When either Program 7 or 8 is selected, the NanoSpec's wavelength counter display changes from 480 to 520 nm. A new reference check must be run on a bare silicon wafer. The reference measurement takes data at 520 nm for about 3 seconds, and stores it in memory.

Note: Critical Focusing for Optimum Results on Thin Films Programs 7 and 8 require very careful focusing on the wafer surface.

To measure a sample, the instrument takes reflectance data for the sample at 520 nm and compares it with the reflectance of the bare silicon reference. Film thickness is calculated from this comparison.

When measuring very thin films in this way, the calculation depends strongly on the reflectivity from the bare silicon. For this reason, use silicon from the same manufacturer with the same crystal orientation, doping and so on, as that used in the samples. Although these programs are specified up to 800 Å Programs 1 and 2 are better for films above 500 Å. Because data are taken at a single wavelength, this algorithm can mistake a thick film for a thin film of measurements, measure only films that are known to be less than approximately 800 Å on oxide and less than approximately 600 Å on nitride. If unsure of the correct thickness, check the range with Program 1 or 2.

When the computer calculates a thickness less than 20 Å, the message: LESS THAN 20 Å appears. For increasing confidence in measurements of very thin films, repeat the measurements. For very thin films, a new reference should be run at least every thirty minutes.

9. Polyimide on Silicon

Measurements made with this program are in the range 500 - 30,000 Å.

10. Positive Resist on Silicon

Measurements made with this program are in the range 500 - 40,000 Å. The refractive index (RI) for the particular positive resist must be known. The default RI is 1.63.

11. Positive Resist on Oxide

The sum of the positive resist thickness and the underlaying silicon dioxide thickness must not exceed 30,000 Å. It is important to enter an accurate, premeasured oxide thickness. Any error in the oxide thickness entered results in an error in the resist thickness measurement. The refractive index (RI) for the particular positive resist must be known. The default RI is 1.63. The instrument will display "Under-range" when measuring oxide of less than 10 Å.

12. Reflectance Mode

The reflectance program, unlike the direct film thickness programs, measures the amount of light reflected from a sample at up to three wavelengths specified by the user. Repeatability is ± 2%. The wavelengths to be specified depend on the type of light to which the sample will be subjected in later processing. For example, in a process using mercury vapor lamps to expose photolithographic resist, wavelengths of 405, 436 and 546 nm should be specified. Mercury vapor emits strongly at these wavelengths and the reflectance of the resist is low since it absorbs light rather than reflecting it. If incandescent light is to be used, wavelengths of 410, 530 and 650 nm might be suitable. Such light emits on a continuous spectrum with representative wavelengths at the violet, yellow-green and red portions of the spectrum.

Use program 12 with any lens, however, measuring film under 380 nm requires the UV option with the 15X lens.

When program 12 is specified, the computer requests sample identification. After the operator has focused on the sample, the computer asks the operator for the number of wavelengths to be used (1-3) and then for their values. These are to be specified by a three-digit number in the range 200 - 800 nm.

Measurement time may be varied between 0.1 second and 6.8 seconds in steps of 0.1 seconds. Repeatability is substantially improved when measurement times are longer than 3 seconds. Longer times can be used at any wavelength but are particularly recommended when measuring wavelength below 400 nm.

When a measurement is taken with this program, the reference reflectance at each specified wavelength is measured and stored.

Attention: If the yellow filter is removed, be sure to replace it after completing reflectance measurements.

If results of the reflectance measurements are question- able for any reason, better results may be obtained by varying the standard measurement time of 2.5 seconds for program 12. For example, if one or more of the wavelengths is 400 nm or less, the accuracy of the measurement may be reduced by the low signal-to-noise ratio at these short wavelengths. To vary the measurement time, request program 120 instead of 12; the two programs are identical, except that 120 asks for a sample time, which can be between 0.1 - 25.5 seconds, in steps of 0.1 second. The reflectance is measured for the length of time specified. Repeatability of the measurements is substantially improved with measurement times longer than 3 seconds. Longer times can be used at any wavelength but are recommended below 400 nm.

Dopants in the film tend to change its color, and therefore its reflectance at particular wavelengths. Such factors should be taken into account in evaluating the results of reflectance measurements.

13. Thick Films

This program is for measurement on thick films (4 – 25 μm). It scans the sample from 720 -760 nm at a slower speed. The operator should be told that this slower scan speed will make a different sound, which does not indicate a malfunction. Best results are obtained with the 5X lens; however, the 10X lens may be used with films of lower refractive index value in the thinner portion of the thickness range. The instrument will display "Under-range" when measuring thick films of less than 0.01μm.

14. EPA914EZ Resist on Silicon

This program was specifically designed for use with Mac Dermid's EPA914-EZ positive dyed resist. The instrument will display "Under Range" when measuring thicknesses less than 100Å.

15. Very Thin Oxide on Silicon

Not available at this time.

16. Very Thin Oxide on Silicon

Not available at this time.

17. UV Thin Oxide on Silicon

Program 17 measures very thin silicon dioxide on silicon in the range of 25 – 175 Å. To ensure accurate measurements, use an ellipsometer calibrated 0- 60 A oxide sample.

18. UV Thin Nitride on Silicon

Program 18 measures very thin nitride on silicon in the range of 25 – 500 Å. To ensure accurate measurements, use a bare silicon reference wafer.

19. UV Oxide on Polysilicon

Program 19 measures silicon dioxide on polysilicon in the range of 150 – 10 kÅ. This program reduces measurement errors, caused by polysilicon scattering the UV light by requiring two reference scans. The 1st reference scan is made on a bare silicon reference wafer. The 2nd reference scan is made on a sample of your bare polysilicon on oxide, or an oxide on polysilicon sample where the oxide thickness is known.

20. UV Oxide on Aluminum

Program 20 may be used to measure silicon dioxide on aluminum in the range 500A-20 KA. This program reduces measurement errors, caused by aluminum scattering the UV light, by requiring two reference scans. The 1st reference scan is made on a bare silicon reference wafer. The 2nd reference scan is made on a sample of your bare aluminum, or an oxide on aluminum sample where the oxide thickness is known.

Note: Programs 17 to 20 require the DUV lamp be turned on 15 minutes prior to any measurement (including the reference wafer). A reference must be taken every time a UV test is run also.

21. Oxi-Nitride on Silicon

Program 21 measures the thickness of a single layer Oxi-Nitride