Why "Astonishing"? What is a measurement?
In the preprint “The impact of a measurement on an open quantum system” we state: “The impact of a measurement on the above-described system is astonishing.” On this site, we would like to point out what it is that makes these measurements astonishing.
In physics a measurement can be very different from our experiences in everyday life. We may think that we as an observer or experimenter are completely independent from the outcome of the experiment, at an elementary level this is not the case. Somehow, we are linked to the experiment in a way we do not completely understand. This is known as “the measurement problem”. There exist many types of measurements. Below we show an electrical measurement on a molecule squeezed between two pointy electrodes, similar as drawn on the HOME page. This data is obtained by setting (is forcing) the bias voltage (V) for one hundred different values and measuring the resulting current (I) for each value. This provides us with one hundred V and I point-pairs. When these points are drawn in the IV plane it results in the below indicated IV curve.
An IV curve can be obtained by setting many bias voltage values and measuring the corresponding current values, leading to the shown IV curve. In this experiment the voltage was set one hundred times in the 7.5-10 V range (the graph shows only a section of this range). There seems to be a clear smooth curve up to 8.3 V and a noisier curve from 8.3-9 V.
In our everyday “Independent Experimenter” world we assume there exists a continuous curve, over the entire voltage range, which reflects the absolute truth. Every voltage value corresponds to one current value, independent of the way it is measured, resulting in a “classical continuous curve”. This classical continuous curve can be approached by increasing the number of measurement points. To get more insight into the noisier part of the above graph, the measurement was extended to include one thousand measurement points, see the graph below.
A classical continuous curve is shaping by measuring enough data points. The noisier part in the first graph turns out to be a regular spaced oscillation.
Now we have good visibility on the right side of the graph as well, clear repeated oscillations show, which were missed in the first graph due to a too low number of measurement points. This becomes even clearer in the inset of this graph where the 8.5-9 V range is enlarged. Also, the complete curve seems to be continuous.
Why does all this matter? Below we will show that in an open quantum system the classical continuous curve does not exist, but is replaced by a pattern of data points, where each data point is influenced by the measurement action from the previous data point. The measurement data shown here is obtained from an open quantum system as detailed at the bottom of the HOME page.
Alternating data points measured on an open quantum system provide for a non-characteristic dataset. Increasing the number of data points will not lead to a classical continuous curve as it is the measurement action itself responsible for the observed pattern! Resulting in a completely different behavior from the known classical continuous curves.
Clear alternating measurement points are observed. This implies that when the voltage is increased with discrete steps the measured current shows a high value, low value, high value and so forth. Very different from the measurement curves in the two graphs shown at the top. Could we have missed an oscillating (sinusoidal) classical continuous curve, underlying all data points? An oscillating classical continuous curve where we happened to have measured only the top and bottom points? A curve that was missed because we should have measured with many more data points, similarly as detailed in the two graphs at the top? We cannot go back and increase the measurement frequency in this experiment as devices decay over time, however many more experiments with different measurement frequencies on different but similar devices show corresponding results: alternating measurements.
Because of this there can only be one conclusion, the pattern of alternating measurement points is real! A classical continuous curve does not exist in this case. Each measurement point reflects the impact on the quantum system from the previous measurement. Because a measurement is performed, the quantum system is influenced. This implies that increasing the number of data points will not lead to a classical continuous curve as it is the measurement action which is responsible for the observed pattern! But does quantum mechanics not predict this type of behavior: a measurement cannot be viewed as independent from the system on which the measurement is performed? It does, and quantum mechanics being the most successful theory of all times, undoubtedly the above result will fit in. Nevertheless, even after a lifelong exposure to quantum mechanics and quantum effects it does not prepare oneself for the level of weirdness in the lab, observing your MCB device together with the measurement equipment to start a “one-two” life of their own. Not only making you feel completely redundant as an experimenter, but providing for a “robots are taking over” experience as well. Observing the memory of the single molecule system in action brings it to live. Without a doubt the most astonishing measurement in my lifetime!
