pandas read_sql used for CREATE in Postgres db giving TypeError for SQLite

Question

I am trying to use pandas to create functions and tables in postgres.

When I try this query:

conn = = psycopg2.connect(host='localhost', database='my_data',
                     port='5566', user='postgres', password='postgres')

create_example = '''
    create temporary table example(id int primary key, str text, val integer);
    insert into example values
    (1, 'a', 1),
    (2, 'a', 2),
    (3, 'b', 2);
'''
pd.read_sql(create_example,conn)

I get the following error TypeError: 'NoneType' object is not iterable

The connection/read_sql works for simple queries but I get the error when I try to CREATE a function or table. Why am I getting the error and why does the error seem to be in SQLite when the connection is to a postgres database? tia

UPDATE: Per the observation by @AdrianKlaver I imported sqlalchemy as follows.

from sqlalchemy import create_engine

def db_connect():
    db_connect_string = "postgresql+psycopg2://{user}:{passwd}@{server}:{port}/{db}" \
        .format(user="postgres", passwd="postgres",
    server="localhost", db="my_data", port="5566")
    return create_engine(db_connect_string)

alchemy_conn = db_connect()

So using this alchemy connection:

pd.read_sql(create_example,alchemy_conn.raw_connection())

>>TypeError: 'NoneType' object is not iterable

If I try to use the cursor function I get the error AttributeError: 'psycopg2.extensions.cursor' object has no attribute 'cursor'