White Papers

Advancements in Calibration Techniques and Moisture Measurements

The author has personally been involved in the analysis of gases sealed in semiconductor devices for the past 29 years. His 37 years of experience in the design of mass spectrometers will be reviewed with emphases place on identifying the independent variables with the measurement problem.

The upper and lower limits of package size, sensitivity, gas volume, dynamic range, precision, and accuracy will be explored. Gas transfer curves will be presented for “normal” semiconductor packages, “correlation” packages, and calibrator volumes. These curves will form the bases of a discussion on precision and accuracy as related to the analysis of Hydrogen, moisture and other gases found in Semiconductor Packages.



The analysis of the gas contents (including moisture) of a hermetic structure is one of the more challenging analytical procedures [1]. Many different calibration techniques [2, 3, 4] have been successfully employed, but the use of the multivolume calibrator (TVCV) in conjunction with a general purpose humidifier (GPH) [5] allowed the Mil-Std program to be implemented [6]. As new semiconductor devices evolve, other gases [7] besides moisture have become critical parameters in assessing the potential reliability [8] of a device. Analytical methods [9] beyond those specified in Mil-Std 883D, method 1018.2, Procedure 1 are required to provide accurate information on Hydrogen, Oxygen, and other gases as well as moisture. The following experiment describes a method and procedure for generating gas volumes and mixtures traceable to NIST for the calibration of mass spectrometers used in the analysis of the gas contents of hermetic structures.



Two variations of the (TVCV) have been developed (Table 1) and the original (TVCV) has been expanded to include 10cc, 100cc, and 200cc volumes.



Vol Range in CC 0.001 0.01 0.10

0.005 0.05 0.1

0.01 0.1 1 0.1 0.2-1.0* 10.00



*Variable Volume

Two of the multivolume calibrators are simultaneously connected to the sample chamber of a mass spectrometer (Figure 1).

Gas A




















Pressure Regulator


Two-Pressure Humidity Generator


Multi-Vol Calibrator A


Dew Point Hygrometer


Flow Meter















Sample Chamber


Variable Conductance


Mass Analyzer











Pressure Regulator


Optional Condenser


Multi-Vol Calibrator B


Optional Dew pointer


Flow Meter












Gas B












If you maintain gases A and B at the same temperature and pressure, then we can apply Boyle's Law [10] and the resulting gas mixture is simply given by:

Gas (A) = (VA/(VA + VB)) x 100% (1)

Gas (B) = (VB/(VA + VB)) x 100% (2)

Selecting the larges volume (200cc) for Gas (A) and the smallest volume (0.001cc) for Gas (B) allows us to make a 5.0 ppm mixture. If Gas (A) is humidified, then the resulting moisture will be subjected to the same dilution factor given in Eq(1). Optionally one could vary both temperature and pressure to increase the number of combinations available, however the General Gas Law [11] will have to be applied.

The first experiment consists of using Nitrogen for Gas A and Hydrogen for Gas B. The second experiment uses Room Air for Gas A and Helium for Gas B. Table [II] summarized these test results.






1 5.418E-6 H2/N2 1062 +/- 50 1015 - 43

2 5.418E-6 HE/AIR 4613 +/- 50 4652 - 38

3 1.078E-4 H2/N2 1283 +/- 50 1252 - 42

4 1.078E-4 HE/AIR 4880 +/- 50 4821 - 39

5 1.176E-3 H2/N2 2295 +/- 50 2364 - 41

6 1.176E-3 HE/AIR 5320 +/- 50 5330 - 40

7 1.100E-2 H2/N2 4875 +/- 50 4920 - 40

8 1.100E-2 HE/AIR 8510 +/- 50 8601 - 46

9 1.200E-1 H2/N2 3410 +/- 50 3362 - 41

10 1.200E-1 HE/AIR 7723 +/- 50 7715 – 45


The expected moisture level is the value measured by the Dew Point Hygrometer (Figure 1.) multiplied by the dilution factor (eq.1.). The negative values applied were taken from the calibration data provided by General Eastern for the Hygrometer used in the experiment. The uncertainty of the General Eastern Transfer Standard [12] is given as +/- 0.056 C and includes the NIST uncertainty.

A Princo Nova Barometer was used as a pressure standard. An MKS Model 116A was calibrated using the TVCV and in turn the Baratron was used to determine the volume ratio's given in Table (2).



Many factors influence the data obtained from a Mass Spectrometer and have been reported previously [9]. Since these factors are instrument dependent and unique to a particular design, the dual TVCV calibration method provides a procedure for measuring the sensitivity, linearity, and dynamic range of an instrument for various gas mixture and volumes only requiring pure gases for this assessment.




Ben Moore of Rome Laboratory has provided invaluable insight into this measurement problem and was the source of motivation for these experiments. Frona Wilson produced this paper and provided the funding through Pernicka Corporation for the above experiments.




•  J. Hartley in A Method of Measuring the PPM Moisture – Sensing Limitation of Mass Spectrometers in Testing Small Packages, edited by H. Schafft, S. Ruthberg, and E. Cohen (NBS Special Pub. 400-69, Washington DC, 1981) p. 25.

•  R. Merrett in A Dynamic Method of Calibrating a Mass Spectrometer Used For Measuring the Water Content of Semiconductor Encapsulations, edited by H. Schafft, S. Ruthberg, and E. Cohen (NBS Special Pub. 400-69, Washington DC, 1981) p.33.

•  R. Haack and A. Shumka in Microcircuit Package Gas Analysis, edited by H. Schafft, S. Ruthberg, and E Cohen (NBS Special Pub. 400-69, Washington DC, 1981) p.43.

•  K. Perkins in Three Volume Calibration Valve – Calibration and Operation Procedure for the Carrousel Mass Spectrometer System, edited by E. Cohen and S. Ruthberg, (NBS Special Publication 400-72, Washington 1982) p.8.

•  S. Hasegawa, personal communication.

•  B. Moore in Mass Spectrometer Measurements at RADC, edited by H. Schafft, S. Ruthberg, and E Cohen (NBS Special Pub. 400-69, Washington DC, 1981) p.3.

•  R. Marti, Seal Laboratory Presentation, JEDEC Meeting Jc-13-9504, Sept. 1996.

•  B. Thomas, personal communication

•  J. Pernicka and B. Raby in The Paradox of Moisture Measurement a Modern Tetralogy, edited by E. Cohen and S. Ruthberg (NBS Special Pub. 400-72, Washington DC, 1982) p.3.

•  A. Roth, Vacuum Technology, North Holland, New York, 1976, pp.22-24.

•  A. Roth, Vacuum Technology, North Holland, New York, 1976, pp.25-27.

•  General Eastern, Certificate of Conformance, Form #A40103046G, Feb. 9, 1996.




Back to Top