Energy efficient lighting for the biological clock

For many years, demand for increased energy efficiency has been the main driver of the development and continuous improvement of new light sources. In addition to luminous efficiency, optimal lighting is also characterized by several quality criteria that cannot be assessed in terms of brightness alone. These include beneficial biological effects on circadian rhythm, sleep behavior, cognition, alertness, and overall wellbeing. These biological effects originate in phototransduction, which occurs in the human eye.¹ They are becoming crucial in designing lighting applications, so that it is important to develop guidelines on how to use biologically effective lighting. In particular, a crucial question needs to be answered: how to assess the energy efficiency of a lighting solution with respect to its biological effectiveness?

Sensitivity to brightness during daytime is given by the photopic luminous-efficiency function, V(λ), which is a function of wavelength (λ). The biological effects of light on humans are represented by the so-called Brainard results (see Figure 1). V(λ) peaks in green, at a wavelength of ∼555nm, while Brainard's results peak in the blue wavelength range, at around 460nm. This difference renders well-known photometric parameters (such as luminous flux or illumination levels) unsuitable for assessment of lighting with respect to its biological effects. Therefore, the unit of lumen per Watt, used for measuring luminous efficacy, is not suitable for judging energy efficiency in this context.

In collaboration with several scientific groups, we have established a metric for comparing the biological effects of light sources. We defined a circadian function, c(λ) (see Figure 1).² The lamp spectra are first weighted ‘biologically’ against c(λ) and then compared to spectra weighted with respect to V(λ). The ratio of both scores is defined as the circadian factor. In the Lighting Technology Standards Committee FNL 27 (Effects of light in Humans) at the German Institute for Standardization (DIN), we have overseen incorporation of c(λ) into the German pre-standard DIN V 5031-100.³This enables a qualitative comparison of the biological effectiveness of light sources.

Figure 1. Results of Brainard¹ and circadian function,²c(λ), compared to the photopic luminous-efficiency function, V(λ).

However, appropriate units for quantification of the biological effects of lighting are still lacking. We must define a ‘biologically’ rated lumen, which would be analogous to the ‘scotopic lumen’ used for night vision. The latter is based on the scotopic luminous-efficiency function, V ′ (λ). V ′ (λ) peaks at 1699lm/W, V(λ)at 683lm/W, and c(λ) at 12,418lm/W. Such a large peak value cannot be used to compare visual and biological effects: it could lead to confusion similar to that affecting comparisons of photopic and scotopic illumination levels for mesopic vision.

Instead, a more realistic approach uses c(λ) and simultaneously introduces a new coefficient for multiplication of the biologically rated spectral quantity. This coefficient may be derived from natural daylight, which is generally accepted as a reference for biologically effective light. Using the standard illuminant D65 (see Figure 2), a coefficient may be defined such that its evaluation gives equal values for the visually rated quantity (e.g., luminous flux in units of lumen) and the new biologically rated parameter (e.g., biological flux in ‘biolumen’). In daylight, with a total luminous flux of 1000lm, this biologically rated coefficient reaches 17,110lm. However, to obtain an equivalent flux in biolumen (biolm), the coefficient needs to be corrected by multiplying it by 1/17.11biolm/lm. This delivers a final value for the biological flux of 1000biolm. If we then divide the peak value of c(λ) by this correction factor, the maximum biological efficiency of light would be reached at 725.8biolm/W.

Figure 2. Daylight spectrum created with the standard illuminant D65. rel.: Relative.

This definition allows for simultaneous comparative assessment of light sources and lighting solutions with respect to their photopic and biological energy efficiencies. Table 1 shows representative quantities for light sources used in general lighting. While for daytime lighting both photopic and biological energy efficiencies should have high values, for nighttime lighting the biologically rated energy efficiency should be minimized.

Although the biolumen has not yet been defined properly, it is clear that its use could help balance the photopic and biological effects of light, without contradictions as regards overall efficiency at high energy demands. At present, the FNL 27 group is discussing standardization of the biolumen as a unit. A second edition of the German pre-standard DIN 5031-100³ will be published this year.