Purpose

This document describes the diode mode quantum efficiency (QE) measurement process at the University of Arizona Imaging Technology Laboratory. Diode mode QE testing is the process of measuring the photocurrent generated from an entire detector (such as a CCD) and comparing it to the photocurrent generated by a standard diode. The detector is not powered on for this test. In the case of a CCD, the photocurrent is measured across the buried channel, typically using the RD and SUB connections.

Equipment

Software: National Instruments LabVIEW and Microsoft Excel
Data acquisition board: National Instruments PCI-GPIB IEEE 488
Photodiode: United Detector Technology, UV enhanced, PIN 10321, last calibrated from 200 – 1200 nm on 8/20/2002
Keithley Electrometer 6512
For QE measurements with l ≥ 300 nm (all Oriel Instruments part numbers):

Integrating sphere: Labsphere 20" diameter, IAS-200 SF

Light source: 150W Xe ozone free lamp (6255)

Arc lamp housing (66002)

Universal power supply (68805)

Light intensity controller system (68850)

MultiSpec 257 monochromator (77702)

quadruple grating turret (77708)

1 mm x 15 mm fixed slits (77733)

Motorized filter wheels (77737)

600 l/mm 200nm blaze grating (77743)

600 l/mm ruled grating (77744)

For QE measurements with l < 300 nm (all Oriel):

Integrating sphere: Labsphere 4" diameter, US-040-SL CA-02117-000

Light source: 30W D₂ ozone-free lamp (63161)

Deuterium lamp housing (60023)

Deuterium lamp power supply (68840)

Oriel interferences filters from 200 – 300 nm: 53310 (200 nm), 53320 (220 nm), 53330 (240 nm), 53340 (250 nm), 53350 (260 nm), 53355 (270 nm), 53360 (280 nm), 53365 (290 nm), 53370 (300 nm)

Measurement Overview

We place a calibrated diode in our optical beam and take current readings at each wavelength of interest. The diode is then removed and the detector (CCD) is located in the same position. Current readings are taken from the detector at the same wavelengths. The detector may be in either a dewar or in a basic mounting fixture. The values can be corrected for the transmission/reflection loss of any dewar window and for quantum yield in the UV.

A National Instrument’s LabVIEW program is used to acquire and manipulate the current data. The program sets the monochromator to the starting wavelength and then acquires current readings from the electrometer. The program scans over a specified range of wavelengths, acquiring current readings for each wavelength. Typically, scans from 300nm to 400nm have increments of 20nm, and scans from 400nm to 1100nm have increments of 50nm. This is because there are typically rapid QE changes at shorter wavelengths but relatively slow changes at longer wavelengths. The program outputs the median of many current readings for each wavelength in the scan, saving them to a spreadsheet file. This data is used to automatically compute QE using an Excel program.

For l ≥ 300 nm, we use an Oriel Hg-Xe 150 W short arc lamp as an illumination source. A photo-feedback unit stabilizes the power supply driving the lamp. Because of this stability (independence from line voltage variations) we do not measure the calibrated photodiode after each wavelength setting but typically once per day.

While the short arc lamp produces more UV radiation than a deuterium lamp, it is not as useful in the UV because of the large visible and IR output. When making UV measurements (l < 300 nm), even a few tenths percent visible leak can swamp the UV intensity. For this reason, we use a deuterium lamp for UV measurements. Because the flux is very low in the UV when using our large integrating sphere, we have a separate smaller integrating sphere and filter set for the 200 – 300 nm range.

Monochromator and filters

For l ≥ 300 nm, we use an Oriel MultiSpec 257 monochrometer to select the desired wavelength for each test. This unit is computer programmable via a RS232 interface. Two filter wheels are controlled by the monochrometer for neutral density and order blocking selection. The neutral density filters are required to reduce the light intensity to ensure the signal on the CCD does not saturate the device. An exit and entrance slit can be selected to specify a particular bandpass. We typically use 1 mm slits providing about a 100 A bandpass with a 600 line grating.

For l < 300 nm we use an Oriel filter set and a dedicated, smaller integrating sphere. The basic measurement process is the same, but the filter selection is controlled manually rather than through the monochrometer.

Calibrated Diode

We use calibrated photodiodes as standards against which the CCD QE is calculated. The diodes are UV enhanced and have 613 mm² area. The diodes are placed in the same location as the CCD to minimize the calibration requirements of the system. They are calibrated each year and traceable to NIST standards.

Measurement Procedure for CCDs

Turn on optical system bench, including all required lamps and the electrometer. Allow to warm up at least 15 minutes.
Place calibrated diode in measurement fixture and connect to electrometer.
Make sure monochrometer is in remote mode and run a diode scan using the QE Data Acquisition program.
Using the clean bench in the Characterization Lab, place CCD in the appropriate fixture with at least Reset Drain and Substrate wires connected to the BNC QE connector. Record the serial number of the device.
Remove diode from test systems and install CCD. Connect to electrometer.
Run a CCD scan.
Run Excel CCD-QE program. Select device from pull down list to account for the light sensitive region of the detector. Press Open Links to calculate QE from most recent scans.
Press Create QE sheet to store QE data in a separate file. Save this new file as the ITL serial number (snxxxx.xls).
Remove device from test system and uninstall from test fixture on clean bench.
Update ITL database for this device.