Electronic – Is it true that the power line frequency is kept accurate by atomic clocks

I've been curious for awhile now ... and was wondering if there exists components that provide clocks much faster than a typical CPU can, such as up to 10 GHz or higher.

Opto-electronic Oscillators (OEOs) are oscillators that take a photonic signal, like a pump laser, modulate it, and convert it to an electrical signal using a photodiode. The signals generated by these OEOs have extremely high Q factor and thus very low jitter. Here is a diagram of an OEO, taken from this overview of OEOs. The focus here is on ultra-high stability, not a high frequency output. But, there are also OEOs that achieve high frequencies, for instance this dual-loop OEO achieves a tuning range of 32 to 42.7 GHz.

Besides photonic oscillators, frequency synthesizers can provide clocks above 10 GHz. As other answers have mentioned, these can achieve frequencies way above 10 GHz. For instance, Analog Devices makes a frequency synthesizer that generates frequencies up to 13.6 GHz. In addition, synthesizers generate the ferquencies for a signal generators such as this one, which can reach 67 GHz.

Here's a brief overview of synthesizers if you want to read it.

A synthesizer is composed of a PLL (which contains a VCO), and sometimes also a microcontroller as a means of adjusting the PLL digitally.

Quoting from an Analog Devices tutorial on PLLs:

A phase-locked loop is a feedback system combining a VCO and a phase comparator so connected that the oscillator maintains a constant phase angle relative to a reference signal. Phase-locked loops can be used, for example, to generate stable output high frequency signals from a fixed low-frequency signal.

A VCO (Voltage Controlled Oscillator) is a circuit that generates an output frequency controlled by a tuning voltage. One way to implement a VCO is to apply the tuning voltage to varactors, which adjusts the capacitance of the LC tank in the circuit and generates a different frequency.

Basically, a PLL is used to generate an in-phase multiple of a lower reference frequency. They are used to clock data converters, which can go up to multiple GSPS, and CPUs as well.

Besides PLLs, there are a variety of crystal oscillators (TCXOs, OCXOs, Sapphire Oscillators, GPS disciplined Oscillators, etc.). However, unlike synthesizers, they output a fixed frequency. They are usually designed for ultra-low phase noise and long-term stability, not high output frequencies. Due to these characteristics, they are often used as a reference for PLLs.

As with anything that is adjustable, when comparing it and calibrating it against something else that is adjustable, how do you know what is "right"? What is considered "more accurate" if everything is just compared against something else that is compared against something else? There has to be something at the end of the chain where you say "This is exact". Some base value that all others are, ultimately, compared to.

And that is the Caesium atom.

I found a good description of how they work online:

Inside a cesium atomic clock, cesium atoms are funneled down a tube where they pass through radio waves . If this frequency is just right 9,192,631,770 cycles per second then the cesium atoms "resonate" and change their energy state.

A detector at the end of the tube keeps track of the number of cesium atoms reaching it that have changed their energy states. The more finely tuned the radio wave frequency is to 9,192,631,770 cycles per second, the more cesium atoms reach the detector.

The detector feeds information back into the radio wave generator. It synchronizes the frequency of the radio waves with the peak number of cesium atoms striking it.

A good caesium clock has a precision of in the order of \$\frac{1}{3\times10^{15}}\$ which is far better than any crystal, OCXO, TCXO, or otherwise.

So having that as a baseline you can now calibrate other systems against it. And the further systems against those. The higher the accuracy of your nominal frequency you want the "closer" to that source you want to get.

But as has already been mentioned in the comments, that's only half the story. The whole purpose of OCXO or TCXO is not to make a crystal oscillate more closely to that precise "source" reference frequency, but to keep the it oscillating at a fixed frequency. A crystal's resonant frequency drifts and changes depending on temperature. By either controlling the temperature (OCXO) or compensating for the changes in temperature (TCXO) you can either reduce or negate that drift.

Very often it doesn't matter one jot if you are a few Hz out when dealing with MHz or GHz frequencies. What matters is that you stay that same few Hz out and don't drift. It doesn't matter (for instance) that everyone would have to tune their TV to 512.000038MHz (that's what "fine tune" is for), but people would get annoyed if they had to keep re-tuning between 511.999381MHz and 512.000482MHz all the time depending on the weather.