Opto coupler as AC relay trigger?!?!?

[int2str]

I'm looking at the schematics for an electric vehicle "charging" station that uses a contactor to switch the main AC wires. Here's the interesting bit:
It uses 2 MID400 opto-couplers (Datasheet) in different ways (as far as I can tell).

1. It uses one MID400 to sense AC line voltage.
This is exactly what it was designed for. Easy.
AC line voltage is connected to the Vin1/Vin2 side (the emitter) and the DC output signal is hooked up to the microcontroller. AC in, DC out.

2. It seems to use another MID400 in an inverted way to SWITCH the contactor on that ultimately powers the cars charger.
So, if I understand the schematic correctly, a microcontroller pin (5V) is hooked up to the input terminal (i.e. the emitter). The AC Line is hooked up to Vcc and Vo (in line). So DC in, AC out?!?!

Is this a common application for opto-couplers?
Why/how does this work?
Assuming the contactor coil probably draws a bunch of mA, is this safe?

[max666]

No this can't be, the AC line can never be connected to the detector side (Vcc, Vo or GND).

But you can use the MID400 as a normal optocoupler and connect a microcontroller output pin to Vin1/Vin2 (emitter) with a now smaller series resistor. The Detector side however can only be connected to 7 V tops and Vo can only sink up to 20 mA (so probably not even enough for a normal relay). Check if the "contactor" that ultimately powers the cars charger is a solid state relay, that's the only way that would make sense for me at the moment.

[int2str]

Makes sense, max. Thanks!

Turns out the 2nd chip is NOT an MID400, but is indeed a solid state relay (AQH1213; same DIP-8 package).
The part wasn't labelled in the schematics and the pictures weren't clear...

Mystery solved 

[eneuro]

It uses 2 MID400 opto-couplers (Datasheet) in different ways (as far as I can tell).

Even 2 cheap PC817 input diodes can be put in anti parallel configuration and with 10k pullup on its outputs you can detect as a bonus positive and negative parts of AC mains which probably is not possible using this MID400 while you have only one output, so we do not know which input diode conducts
So, in spot welder project latest upgrade I used such PC817s to wait in its control software to switch transformer always at the same part of  AC cycle for full wavelength periods of time, so transformer primary coil is always demagnetized by previous runs and if you do not reverse load and neutral lines when pugging into wall socket, than control circuit monitors AC wave form and can again send pulses to trasformer  when AC voltage has polarity that will not run into saturated transformer core 

Before playing with PC817 input diodes, made such simple experiment with common LEDs and they survived 230VAC mains

So, I've canceled order of this H11AA1Optocoupler, Phototransistor Output, AC Input
 http://www.vishay.com/docs/83608/h11aa1.pdf