Q1IC Technology

Resistivity (denoted by the Greek letter rho) is an intrinsic material property that quantifies how strongly a material opposes the flow of electric current, independent of the specific dimensions of a particular sample; it is related to the measured resistance R of a uniform conducting bar of length L and cross-sectional area A by the relation R = rhoL/A, so that resistivity has units of ohm-meters (or, commonly in semiconductor work, ohm-centimeters). In a semiconductor such as silicon, resistivity is determined by the free carrier concentration (electrons and/or holes) and their mobility, through the relation rho = 1/(q(nmu_n + pmu_p)), where q is the electronic charge, n and p are the electron and hole concentrations, and mu_n and mu_p are their respective mobilities; since carrier concentration in a doped semiconductor is set almost entirely by the net ionized dopant concentration, resistivity measurement is one of the most important and widely used techniques for characterizing the doping level of a processed silicon wafer.

Van der Pauw Technique for Irregular Sample Resistivity Measurement

The standard four-point-probe technique for resistivity measurement assumes a specific, well-defined sample geometry (typically an infinite or semi-infinite flat sample with four collinear probes at fixed spacing), which makes it unsuitable for samples of arbitrary or irregular shape. The Van der Pauw method was developed specifically to overcome this limitation, allowing accurate sheet resistivity measurement on a sample of essentially any shape, provided the sample is of uniform thickness, is singly connected (contains no isolated holes), and the four ohmic contacts are placed at arbitrary points on the sample periphery, with contacts made as small as possible and as close to the edge as practical to minimize measurement error. The technique involves passing a known current between one pair of adjacent contacts (say contacts 1 and 2) and measuring the resulting voltage across the other pair (contacts 3 and 4), yielding one resistance value R_12,34 = V_34/I_12; the measurement is then repeated with current passed between a different pair of adjacent contacts (2 and 3) while measuring the voltage across the remaining pair (4 and 1), yielding a second resistance value R_23,41. Van der Pauw's theorem relates these two resistance values to the sheet resistance Rs of the sample through the transcendental equation exp(-piR_12,34/Rs) + exp(-piR_23,41/Rs) = 1, which must generally be solved numerically or graphically for Rs, although for the special case where the sample geometry gives R_12,34 = R_23,41 (a symmetric configuration), the equation simplifies to Rs = (piR_12,34)/ln(2). In practice, both current directions are also reversed and averaged for each measurement to eliminate offset voltages due to thermoelectric effects or contact asymmetries, and the resulting sheet resistance can then be converted to bulk resistivity if the sample thickness is known, via rho = Rst.

Thin Film Resistor Sheet Resistance Calculation

The resistor must dissipate a total power P = 5W when connected across a source voltage V = 12V. Using the power relation P = V^2/R, the total resistance required is R = V^2/P = (12)^2/5 = 144/5 = 28.8 ohms.

The sheet resistance Rs of a thin film resistor relates to the total resistance R through the number of unit squares the resistor pattern comprises, defined as N = L/W where L is the total length of current flow through the meandering pattern and W is the uniform trace width, since each square segment of the film (where length equals width) contributes exactly Rs ohms to the total resistance in series, regardless of the absolute size of that square. Here, L = 5m and W = 1mm = 0.001m, giving the number of squares as N = L/W = 5/0.001 = 5000 squares.

Since the total resistance is the sheet resistance multiplied by the number of squares in series, R = Rs*N, the required sheet resistance is Rs = R/N = 28.8/5000 = 0.00576 ohms per square.

This very low required sheet resistance value (well under 6 milliohms per square) indicates that a highly conductive thin film material, such as a metal film (e.g., a thin layer of nichrome, tantalum, or a similar resistive alloy engineered specifically for low sheet resistance at this thickness) rather than a doped semiconductor film, would typically be required to realize such a de-icer heating element in practice, since doped polysilicon or diffused silicon resistor films used elsewhere in IC fabrication generally exhibit sheet resistances several orders of magnitude higher than this value at comparable film thicknesses, and would require an impractically short and wide pattern to achieve such a low total resistance within the given 5000 Angstrom thickness constraint. The stated 5000 Angstrom (500 nanometre) film thickness would, if desired, allow the required bulk resistivity of the film material to be back-calculated via rho = Rst = 5.76e-3 ohm/sq 500e-9 m = 2.88e-9 ohm-meters, a resistivity consistent with a good conducting metal film rather than a semiconductor.

It is also worth noting that the Van der Pauw technique's accuracy depends critically on the contacts being small and placed precisely at the sample periphery, since the underlying mathematical derivation assumes point contacts located exactly on the boundary of an arbitrarily shaped but simply connected (hole-free) lamina of uniform thickness; in practice, finite-sized contacts introduce a small systematic measurement error that increases as the contact size grows relative to the overall sample dimensions, which is why careful sample and contact geometry design (keeping contacts as small as practically achievable, and placing them as close to the true sample edge as possible) is an essential part of obtaining reliable sheet resistance data using this method, particularly for small or irregularly shaped test samples cut from a processed wafer where standard four-point-probe geometry cannot be applied.

It is also worth noting that resistivity measurement forms an essential process control checkpoint throughout wafer fabrication, from verifying that an incoming starting wafer meets its specified resistivity range before any device processing begins, through confirming that a well or diffusion step has achieved its intended dopant concentration, to final electrical characterization of a completed device wafer, meaning both the standard four-point-probe technique for regular geometry samples and the Van der Pauw technique for irregular geometry samples serve as complementary tools within the same broader quality control framework used to catch process deviations before they propagate further downstream in the fabrication sequence.